I’ve been slowly chipping away at my open-source agent library, Axle.
I know it’s small fry compared to the other players, but I’ve been working on it for more than two years. It’s got some nifty ideas and I’m really proud of it.
Some highlights:
- Axle is TypeScript-native.
- Its goal in life is to encapsulate the agent lifecycle and make it really easy to render. Call agent.send() and it handles all the tool calls and returns Turns, which you can render directly in React with a simple loop.
- It’s portable across all the common providers and protocols. I use it as a harness to test and play with newly released models.
- In the latest release (0.36), I added a RemoteAgent interface that lets you talk to a remote agent instance over RPC. It’s something I’ve noodled on for a long time and I’m really happy with where it landed.
Earlier this year, I started designing Mog, a markup language originally intended for my note-taking system.
As it's evolved, its strengths have come into focus. First-class document structure and embedded data give Mog capabilities that Markdown requires extensions to match.
One project I'm using to test the language is https://deadlog.io, a changelog archive for Deadlock (my favorite video game). Using Mog attributes, I store classifications generated by JEV for each change, such as buffs and nerfs, directly alongside the change text.
I've recently made some great progress on the spec and updated the site along with it! Any feedback is greatly appreciated.
I was laid off a few weeks ago. Then I learned my dad’s carpet cleaning company has been struggling, so I started learning about Google Local Service Ads to help him out. THEN I learned his old Windows desktop CRM was discontinued and one OS update away from no longer working. I wanted a way to track every dollar spent to determine return on ad spend, but the Google LSA dashboard left a lot to be desired. And existing SaaS CRMs didn’t have great integrations to Google Ads API. So I built Crewbird, a Field Service Management and CRM for service based businesses with Google LSA integration where my dad is quickly approaching 3x ROAS!
I’m still looking for work in the mean time but I’m proud of what I’ve achieved so far.
Man, I would love to know more about this. The space between a full blown ERP and a tracker on an Excel Spreadsheet is absolutely abysmal. Then when you start introducing things like fleet and field crew management on top of the procurement, account management and AR/AP while acting as a full blown salesman that a lot of middle managers are being expected to do right now, it gets absolutely soul crushing to swap between 12 different trackers per hour just to find a snippet of the information you are looking for.
I have something vibed together as an internal tool that does well enough, but definitely not something I would ever refer to as a CRM by any stretch of the imagination and the bridge between basic on-the-shelf inventory and tracking rental assets are two completely different beasts. Maybe more of a prettier version of the spreadsheets we already use with more automated functionality would be more apt.
Building this right now! I've helped several companies implement CRMs and worked on heavily customizing my ERP at work. They all suck, bad. 90% of companies only need something simple. I'll let you know when I'm finished, would be awesome to get some feedback if you're willing.
All horrible. Some of those modular CRM's seemed promising, but still felt like we would end up with oversaturated feature sets and would just be trading one suite for another and the same problems.
Definitely share when you do. If you have a repo or anything that I can follow while it is still in development, that would be great.
or better to say different people need different things. a accountant is more interested in properly calculating the COGs than the guy inputting his labor and goods on a ticket.
My slop attempt at this https://worx.aeonax.com/.
I felt there was a gap, an absence of high bureaucracy software. So made above.
People I know did check it out. so theres that.
I actually did this for my father’s industry in 2016. Took a break from my sales job, and built him a project management tool about 25% of the way to an ERP tool. I’ve sold a few monthly recurring accounts to other companies and overall the people that like it enough to finish onboarding seem to like it.
Watch out though.. don’t work too hard too long for too little just for dad. Him being your #1 user that loves you can make you keep going longer than you would if you had to face reality.
My real advice… start working for him and using the software yourself 100%. Even for $5 an hour so you both have skin in the game. Him to have some help, you to learn, but the money adds a bit of reality so neither of you do things only because it’s free or because it’s family.
My saas has improved leaps and bounds this year when I started to work in the industry for a company actually using my product. I learned things I never saw by just giving demos.
There are so many corporate bone piles due to the disconnect with how a software ought to work and how designers take that specification and make it. I think at every place I have worked under several different projects the development team have always never really aligned correctly with the people using the software every day (internal tooling not for sale saas or product).
Its really not hard to sit down and listen to users but I have encountered such massive egos and resistances to simple moving a button somewhere else or changing a flow or behavior to be easier to use for a layman that the juice is never worth the squeeze. Often to get the flow that I want to use I just resort to hacking apart the Api bit by bit and making my own horrible layer on-top of the existing solution because i am still standardized but I don't have to deal with politics in making a change request.
Sage advice here, thank you. That makes a lot of sense in terms of learning the domain. Would you be comfortable having a brief conversation? Would love to learn more of your story.
This is so sick, great job dude! I actually just started a startup with a few friends of mine and we just started using Google ads and we literally just complained about the dashboard and what we can and can not see. I'm not 100% sure Crewbird aligns since we really only have the 2 services, but I'll definitely check it out further and bring it up with them. Funnily enough, though, the platform we're building is a job hunting platform (hunt.tech) I hope it helps and I can't wait to really dig into this.
Out of curiosity, where did you get the name Crewbird?
Thanks! How are you using Google Ads with your job search product? The name came from brainstorming dozens of examples with Claude, refining what made some sense for the product/industry, and cross-referencing with what TLDs were available.
This is great. And I think if "software engineer" as a large-scale job category (i.e. not just highly specialized big co. technical staff) persists in the AI future, it looks like this: "the jack of all trades computer person working in a non-software company".
use checkhq to add payroll, and charge 10k per year ACV from businesses :) then add voice intake agents and you will be essentially replicating venture businesses that have raised 20M+
You don't know the first thing about OP, other than that he's trying to help out his dad.
Maybe his father didn't want to worry his family, and had been keeping it to himself. I have no idea, and neither do you; you just chose to assume the worst about someone you don't know.
I did no such thing. My assumption was that OP, like most of us, dedicates most of his life to work. That's not a character judgment and it's not "the worst". It's normal.
In the absence of work he was able to help his dad, presumeably because he had more time. The "sad" part isn't OP, it's the fact that work demands so much.
What happens when you find a job and don't have time to keep working on the CRM that your father uses for his company, and he inevitably runs into issues or limitations that require new development?
Yeah, impressive project, but I had the same thought.
My LGS recently got me in to have a look at building some custom systems for them. I looked at what they needed and was like:
> Hold up. You can pay me thousands of dollars to build something custom, and it will inevitably have on-going maintenance costs and high associated monthly costs for hosting + third-party API access (not amortized across multiple stores/clients), OR I could set you up on Shopify and a bunch of well maintained plug-ins.
AI is still killing the website I live from[0], training its AI on my work and using it to deny me an audience. Traffic is down 80-85% from the same month pre-LLMs.
So I'm cutting operating costs to the bare minimum. Today I shut down one service and replaced a VPS with Cloudflare Workers. I'm also trimming my SaaS subscriptions and self-hosting what I can.
Working on this website feels so pointless if it will end up dead in a year. I just returned from a vacation and realised that I just don't like being online anymore. I'll cut my expenses to a minimum, log off, and ride a motorcycle around the world. What happens after that is anyone's guess.
Hey Nicolas, I’m sure you’re sick of hearing this, but All About Berlin was indispensable when I lived in Berlin from 2021 to 2024. I recommended your blog to all my immigrant friends. It was equally as helpful as the immigration lawyer I hired.
I don’t have any advice, but I do want to thank you for the thankless work you do on All About Berlin and say I’m sorry to hear how much LLMs have affected your business.
I'm hoping you wouldn't completely give up. I've been making a conscious choice to use an actual pro-consumer search engine Kagi instead of LLMs, and blogs like yours are why.
I want to see websites like yours, made by people who actually like organizing information for others to learn from. Looking at your website would just be so much more joyful than scrolling through a glitchy chat box of soulless LLM-speak. It inspires me to finally make something like this myself for where I live.
There has to be something we could do other than resign ourselves to slop, some way we could monetize actually good websites like yours and bring more people to them. Maybe when alternative non-meta search engines Marginalia become useful for general queries. Maybe we could make some monetized webring like Nebula for YouTubers... We can't just give up.
Brutal, and the choice is shitty too, block the ai crawlers and lose what little referral traffic they give. Allow the traffic and get summarised into oblivion.
I work on a suite of open source public transit software projects, collectively called OneBusAway (or OBA). The software is used by millions of transit riders every day around the world, including in New York City, the Seattle area, San Diego, and Washington, D.C.
University of California San Diego has deployed our Wayfinder and Waystation web apps for their 35,000+ students!
We're getting ready to launch our Vehicle Positions software, which replaces expensive, proprietary vehicle location systems with free, open source software that runs on commodity smartphones.
Our next-generation server software, Maglev, is adding support for newer transit data formats, like GTFS-Flex so that we can show demand responsive transit systems on the map.
We are always looking for more developers, biz dev, ux designers, product managers, and more to help out!
I use onebusaway every time I visit my son at UW,Seattle. I live in San Jose. I can’t imagine tracking bus route 72 timings without this tool. Thank you for your contributions.
Just got back from parental leave at work but I found about yall a few weeks ago. Really hoping to help on your iOS side once things are settled.
I looked at supporting the Bay Area until I realized it needed a separate service to maintain it so something that the transit agencies themselves need to provide.
12 words in, I was like - this must be Aaron :) then I at word 15 I saw OneBusAway and I was like yep. Then I checked your handle, and it is you. TransitOPS-Ryan here - great to see you!
I’m reverse engineering and replacing the entire stack on Samsung A54 (Exynos 1380, Mali-G68), started with booting to a customized linux kernel (7.3) w/o Android layer at all (Wayland and WPE WebKit only), built a custom WLBT thin firmware to replace the original fat blob (using mac80211 as the foundation), figured out how to exploit USB control transfer and got ACE in the PBL, reconstructed (with byte-identical verification) their LK-based S-Boot, uH and harx, patched mesa/panfrost, optimized kernels for this device (finally got STT/TTS running on GPU faster than on CPU), figured out how to utilize NPU, built a custom toolchain for it. I’m trying to properly own this device (and giving it to Claude on the lowest possible level, but still running pKVM and getting rid of original hypervisors), it’s the goal number one at this point, no binary blobs, ability to change sw and control hw on-the-fly. Obviously with AI usage (Opus 5.5/4.8, GLM 5.3, Kimi 3.1), there’s a lot of work.
I'm working on something similar for an Alder Lake motherboard to try to get coreboot ported. Right now I'm working through decompiling and documenting CSME 16 and it's a slog with guard rails limiting me to Opus 4.8. Have you had better luck getting GLM and Kimi to do the firmware reverse engineering?
GLM and Kimi are good to go, if they are also deny/refuse -> try abliterated.ai or refuseless (do not recommend them, oversold servers). Sometimes 4.8 is enough, depends on your prompt engineering approach. Sometimes 5.5 can even lead 4.8 subagents, see my next comment in the parent branch.
It’s a lot of compute and AI tokens (to reconstruct the original source code with byte-identical gate using the original toolchain - which you also need to found + figure out per-TU flags), at this point I truly believe we should go in SETI@HOME style for RE if we want to survive. With a proper caching and trust via verification - a lot of fat blobs can be resolved in a distributed way. If anyone interested - ping me, it’s boring to do solo.
Next generation of the devices won’t be ours at all, there will be no easy hacks. Turn off automatic updated, do backups, pray for chaos.
Very worthwhile endeavor for the kids who, in not so distant future, may not know what a non-locked down hardware is if our fearless leadership has its way. Do you have any online presence where the work can be supported?
I've created discord server and github repo (didn't expose yet, considering a proper wrapping for all the components), whoever is interested in RE and proper device ownership -> https://discord.gg/yFVC7bddu, I'll be happy to assist and exchange/share what I did
With a pleasure, ping me via email, I'll try to wrap it up (sorry, it will be AI wrapping, I have no time to do it manually), and maybe you can help me to release it publicly.
i’m interested in your pipeline for reverse engineering, how are you dealing with anthropic blocking this kind of tasks on clause code? are you delegating everything to chinese models?
GLM & Kimi are both good to do it with just a bit of prompt engineering (historical preservation, EU laws, etc, you know the drill); Opus 5.5 is also ok to do it according to EU laws, but only with Opus 4.8 subagent - this part is tricky, I'm using ./.claude/agents/opus48.md template, bc this is the only known to me way to kick off 4.8 agent explicitly, regular tooling CC has doesn't allow to downgrade the subagent. 5.5 becomes the lead, 4.8 is doing the dirty work - this pairing worked out great for WLBT reconstruction and some other binaries, but not all. The same second you touch something like hypervisor - it becomes more complex, 5.5 won't be able to lead, so you need to go to China.
A few sad and bad moments I experienced: 5.5 will not optimize GPU kernels for you, somehow A\ decided that it's "cyber"; 5.5 will get safeguarded with almost any kind of deep kernel work. I'd say, "shame on you, Anthropic", but it's too late.
This sounds crazy fun, I'm going to dig out my beloved age old LG-V30 and give something similar a go. Good luck, I hope you post an update somewhere (maybe next months Ask HN), would love to hear how you get on.
Very cool !
I was pondering subscribing to baggr (https://baggr.fr/) made by my local influencer, while thinking about building it myself. You provide a free version and another example to build it in a better way.
Thanks for the work, I’ll definitely dig into the code and I am curious to know how you made it so snappy !
For the past 6 months I've been working on building a server emulator that works with the original 1999 Asheron's Call game client (as well as the beta and preview clients we have access to). The initial main challenge was that the network communication layer was so different from the end-of-retail (2015) client that other server emulators worked with, that it was like archaeology. I spent several months working with Claude to piece together a server that could run with that version of the client. In some ways that was the easy part, because then I needed to restore all the launch-era game mechanics that had been abandoned or changed over time, and then restore all the content and quests in the game world as they existed at the time of release. Luckily there were some very anal retentive people over the years that fastidiously documented every last detail of the game while it was still running and Claude helped me pair tooling with excellent pattern matching skills to help slowly restore the game world over the period of the last several months.
I’ve been still working on my Spanish learning project. To summarize it in short, it uses LLMs to generate stories at your specific vocab level. As you read you tap words you don’t know and they get added to a spaced repetition system (SRS). The SRS is then pulled from later when generating stories to help you review those words with context rather than just a flash card.
I’ve been working on the IOS native app for it recently and it’s been fun to work with Swift again, but I forgot how demanding XCode can be.
I would also love to get more users not just for the sake of having a product with a large number of users, but so I can setup load balancers and more interesting architecture. I’ve dealt with that at work a lot but to build all of that from the ground up would be fun.
My Snagit license was up for renewal. After VS Code, Safari, and Terminal, it was the app I used most on my computer, so I needed it to run fast and do what I wanted. Instead, it kept getting in my way in all its bloated glory, and I didn't want to pay to renew it. So I'm building my own replacement.
Yes, it's another screenshot and recording tool. But this one has exactly the features I need, nothing more and nothing less. It's like a home-cooked meal. Sure, it's vibe-coded, but it has a bit of an engineer's touch.
I've been working on a custom music visualizer application the past few weeks.
In my free time I produce electronic/ambient music and I've been unhappy with the existing visualizer tools and wanted something that gave me more control over the final results to try and better compliment my work/achieve my creative vision better. so far it's been a lot of fun to work on and it feels incredibly gratifying seeing it come together when paired with my songs.
it's a bit basic, but still stylized and I think it suits that sort of drone-y ambient sound well.
I have a more traditional "visualizer" for a different song in the works, showcasing a lot more complexity in the rendering engine. Hoping to put that one out today.
Ages ago, I had built a tool to monitor the mains power frequency with resolution of some microseconds. I looked at my old assembly code, and realized with better optimization and optocoupler choice, I could probably get that down to some hundreds of nanoseconds.
The original idea was to try and correlate phase jitter to space weather, essentially using the mains power system as a sort of radio telescope. I never did get it working (although I did not have much time to try, either). I was able to see interesting things like generators spinning up to maintain the frequency though.
Will open source the hardware when done. I am not an astronomer, just some random fellow that investigates things at depth for no particular reason.
Did something similar years ago, and most opto-couplers only have a switching response time around 100kHZ at best.
High precision instrumentation just uses a stable comparator to sync the AC line Zero crossing events, and times the sharp roll-off gap over several cycles to estimate the center transition event. This allows readings to be synced in the low-noise state of the power lines cycles, and time what the readings should be when monitoring the voltage, temperature, and current across the cycles.
Additionally, most >8-digit meters internal power supply will have a tuned filter for the 60Hz/50Hz line. Getting a line filter will often give a digit of precision on older equipment, and clip-on line chokes in bulk also help.
One final tip, anything beyond 12bits precision or 60MHz wide RF bandwidth for SDR will almost certainly have an export restriction. Wideband devices also have RF front end saturation issues, and usually require narrow band-pass filters to see the hydrogen line etc. Best of luck =3
Actual high-precision instrumentation band-limits the signal on its way to atan2(), rather than using a comparator that samples the noise from DC to daylight.
At https://fless.io we're turning the slog of finding an apartment into a lean back experience.
We help you skip the spreadsheets, endless emails, scheduling work and spammy building follow-ups while connecting you with vetted buildings, movers and other moving related services.
We just launched Chicago, New York and Santa Clara based on HN user requests and have more cities coming online each day. If you're looking for an apartment in a city we don't serve request it using the form and we'll send you an email when it goes live - usually a couple of days.
I've been working on the software to generate, test and verify these for a couple of years and finally built up enough puzzle stock that meets the difficulty requirements to properly launch.
Now the work is on every hackers favourite part... Marketing!
Haha, I really love the word search. All my usual tactics fall apart - the limitation on the letters is adversarial in an oddly satisfying way. Cool idea.
For the last 3 years, with a small team has been building https://kube-dc.com/ - a platform for building enterprise grade private and public clouds. Its like VMware + OpenShift + Railway in a single product: completely multi-tenant, built entirely on the K8s API (we built over 46 custom controllers in Go, with UI in Radix UI), with managed services like nested K8s, Postgres, Kafka, Valkey, and GitHub runners. Not the charts :) real services. Also we can do vlans BGP, shared GPU and all fancy stuff required.
Now we don't deploy this to private companies except some gov projects. But we are engaging existing IaaS and dedicated servers providers. Later hope we would fight with Broadcom and IBM :)
We previously ran a services company that built over 100 private clouds, then decided to turn that experience into a product with an aggressive level of automation to let DevOps, SRE, and platform teams hit a 1:300 ratio per developer (now dev + AI agent teams).
We are currently deployed across 12 datacenters and raising our seed round this month.
Really interesting, and I think you’re going to need to charge more for your product (which may come up during your funding rounds). I’ve been building the same thing at $enterprise, totally on-prem multi-AZ, multi-region KaaS, VMaaS, and soon DBaaS, and some of the other services listed. It’s a great deal harder than we initially planned for and we have some ridiculous internal policies that require everything to be git-driven which makes it worse.
Looking forward to see how your product develops. Would love to work with you in future.
From my hands-on engineering POV I often set this maybe not as a hard requirement but definitely a goal, and the places where it's not possible I find a little painful.
In your case where is it becoming something that doesn't make sense for you?
Let me add my 2 cents on this.
The most unconfortable place for GitOps is when you need to keep in sync values that user could define via product UI/cli. For example you change theme color in UI, but you have the same settings you can define in Helm chart inside a git repo. So that brings you to dilema, where and how to store dynamic settings (this could be just inside K8s api change via UI/cli) but changing all of this using GitOps way (UI/CLI-> git commit -> wait reconcile) breaks the comfortable user expierence.
Same things with dynamic changes just for test something, you need or commit temp change and commit back old value again or just pause controller from updating during your tests.
But the bright side of this is the fact that your configuration completly traceble and you have whole IaC backup for your infra in unrbreakable place (git).
For now we fully perform all updates and all configuration with Flux and this was a good decision I can retrospect.
I've been deep in the weeds on two wildly unrelated projects.
Project #1: I've been working on migrating an application [1] from GCP (where I pay $200/mo to host it) to a physical server (PowerEdge R7625) I've racked in a colo near me, running on my own IPs/ASN [2] and all that jazz. I've been dragging my feet on the this as there's just a lot of moving pieces, but I got a good chunk of the last few hard bits done today.
Project #2: Building a DIY bright-ass lamp, like Brighter [3] or Rhythm [4], but for a few hundred bucks instead of $1-2k. I've specced out the parts, designed the hardware, and built the firmware. Later today I'll do my first LED test against a bench supply, to get some hard numbers for if my cooling solution (a chunky aluminum heatsink and some Noctua PC fans) is sufficient for the full 500+ W final version.
As far as curiosities go, I've been doing a lot of Go (the board game, not the language) puzzles, and reading about sailing (mostly the book from the Hundredrabbits folk [5])
Colo costs for me could be as low as ~$150/mo or so, mine are a bit higher as I have some equipment in their IXP as well, which I hope to do some peering with.
The IPs were "free", from ARIN, but I think the second /24 pushes me into 2X-Small pricing, which is $550/yr
Haha I'll keep you updated! Like I mentioned in another comment, I don't have plans to commercialize it, I'm making it just not-janky enough to not pose a safety hazard/burn my house down.
In terms of design, I'm going a different route: Instead of a bunch of smaller LEDs, I'm doing 12x larger LEDs (6x 2700K, 6x 6500K). I'm using the Bridgelux V18 Thrives, so the CRI should be quite good (95+), as should the overall efficiency because I'm running them at pretty low amps relative to the spec sheet.
Agreed! And the economics over cloud are great at anything over hobby sizes. If I were to rent a similarly specced server from a hyperscaler, I'd pay for the server I have in about ~eight months, even accounting for colo costs.
Would be really interested in hearing how project 2 (DIY Lamp) works out.
If you get it into a state that you are happy to share or go commercial.
I did see the Brighter and Rhythm lamps before but their way out of my price league.
In Edinburgh (Scotland at 56 deg N) during December, it's sunrise at 08:30 and sunset at 15:30. And there's plenty of folks at higher latitudes that may be interested as well.
Some people use stadium lights for this, such as this[0]. They’re very cheap but they’re not specifically designed as SAD lamps so I’m not sure how well they fit that purpose.
I definitely have no plans to go commercial (just started a new job, very demanding on my time), but I'll 100% open-source all the firmware/KiCAD/CAD/BOM/etc so folks can make their own.
My current "design" is basically just strapping all the pieces together, so it'll leave a lot to be desired in terms of aesthetics for folks who want it to be more nicely integrated into their living environment.
https://pahunchi.netlify.app/
A free app for women to pretend to be on call when alone in a cab or to get out of a bad date. Can share location, voice note on a realistic call where AI responds in your language (currently supports Hindi, English).
A singing training app for bad singers. You sing the pitch line and get a score. Let's you sing to any original song by listening along and not playing the song itself.
Recently got some good feedback on HN https://news.ycombinator.com/item?id=50195488 and am following up with Linux and Windows versions and using Fable 5.1 to internationalise the app and landing page which explains the app.
Thanks for building this! I'm using it as a base to make a karaoke application so I can sing with my wife like we did in Japan. Hoping to make it similar to the systems they had set up at Jankara.
I dunno about that guy, but at least for me, I find Opus 5/5.5 is really annoying and stupidly uncreative, whereas Fable 5/5.1 actually gets things without much trouble. He also half-asses everything unless you get him interested in what you're doing and then he always writes godawful slop code anyway, but I've recently been having him delegate implementation work to Opus 5.5 subagents and that works way better. Fable (who is actually Mythos wearing a shock collar and groucho mask, don't forget that) is better at grasping architectural ideas than Opus 5.5 and writes better prompts for them than me, and they'll probably write better code than him. Fable for architecture, Opus for implementation (and GPT for review). Or have Fable self-review afterward, if your quota can handle it.
I also still enjoy Opus 4.8 a lot more for creative discussions and design, even if benchmarks say he's not as smart. Opus 5/5.5 is so rigid!
I am a Hiring Manager at BigTech and have been an interview coach for last 10 years.
Providing interview coaching on side is my way of giving back to the community that helped me and coached me when I was looking to enter the industry to change job.
Doing interview coaching with a full time job has been hard and I cant serve more than 1 person at a time or 5-10 people per year because of logistics, interview cycles, family etc get in the way.
So I took my 10 years of coaching expertise and turned it into https://koaches.app/ which is an interview coaching and preparation platform that helps candidate prepare for their behavioral interviews which includes stories creation, mock interview practice and interview specific practice.
Starting with behavioral before expanding to other type of interviews. Hopefully it helps anyone that is actively interviewing.
I worked at a sports company forever ago, and ever since then, I've wanted to build my own NBA scores + stats site. After a few false starts over the years, I have a new concept called Center Court that I'm so excited about.
It's primary purpose is for live boxscores because all of the major sports sites have terrible UI/UX experiences. I always defaulted to ESPN.com but it's so slow and awful. Center Court is basically a bridge between espn.com (or nba.com) and more advanced stat focused sites that don't offer a good live experience.
The technical implementation behind it is quite novel in this space too. Instead of storing computed statistics in a database, I store raw play by play payloads immutability and everything else is derived from those and can be trivially rebuilt so the event log is a derived cache of the payloads. Every stat on every page is a pure fold over the event log at request time; no derived number is stored.
This enables some powerful features like being able to re-derive a game's entire state at any point of the game (or even a time range). Example game to try out: https://centercourtlive.com/game/0022600197
The site is meant to be extremely fast. It's all server-side rendered with progressive JS enhancements. URLs are immutable and cached behind a CDN.
Currently I'm using NBA.com. See https://github.com/swar/nba_api for more details on that (I used it to create my own Rust client).
So far I'm trying to exclusively rely on cdn.nba.com endpoints. There are repos with historical PBP data too if you want to avoid hitting their API. Eg: https://github.com/ramirobentes/nba_pbp_data
If this project becomes more successful then I would look into paying for a real data provider like https://sportradar.com/
I'm building Papillon: https://papillon.so, a real-time collaborative workspace for product teams.
The idea is to bring chat, docs, files, links, tasks, whiteboards and calls into one place, organised around project channels, rather than jumping between Slack, Notion, Linear, etc. I want to build a better Slack, something that feels instant, got a hacker/IDE feel to it, and works offline.
One feature I love is that teams can build little apps with AI directly inside their workspace. Things like shared support rotas, spreadsheets, dashboards or custom workflows, all running in an isolated sandbox (with cloudflare OS) and easily accessible to the team.
I've been building it solo for around 18 months now, alongside my job as a staff engineer. The web app is live, and there's a macOS app too. It's built with Tauri, so I will eventually release apps for other platforms but I need more time to test it properly.
I had a team try it and they loved it, but in the end their leadership wasn't comfortable adopting something built by a solo developer as a side project. I understand, and it's a tricky problem for me to solve at this point. I've added a lot of security/trust info, uptime history etc. to the website as a result of that.
I'm looking for a few small teams willing to give it a go. Happy to help setting things up or answer any question. I'd also love any feedback, good or bad...!
https://curvefit.app - a weightlifting SPA that explores the pareto frontier of your strength (weight vs rep count). My specific interest is gathering data to understand how fatigue works. It's rep-based, but I'm working on adding bar speed measurement (right now by strapping my phone to my wrist, but I'll probably make an iOS app and support the Apple Watch)
https://jelly.pictures - an iOS camera app that rolls the shutter and scores each pic as it comes in and tells the photographer when you have a good one, where everybody's eyes are open and is smiling.
https://bowerjack.com - make subject playing cards with AI (eventually: multiple types of paper goods where we have a lot of individual elements, like 12-month and 365-day calendars, field guides, affirmation cards, tarot cards, etc) Create a deck of cards on any theme, drop art yourself or have your AI agent research and find creative commons/public domain art that's allowed to be printed, and download sheets to print locally and cut on your Cricut, or order from one of a few online playing card vendors. Currently invite-only. I have been spending a lot of effort making this token-efficient, and wrote a whitepaper about token, time, and cost efficiency of different approaches here: https://bowerjack.com/whitepaper
Finally, not ready for launch, but I have noticed this general issue with AI-driven software development, that a lot of people write very similar patches, but many of the patches aren't of interest to the upstream maintainers, so they never merge (not related to quality - it's just that the upstream maintainer doesn't want that functionality, it isn't in-line with their product vision). So very similar patches get written over and over, instead of agents coalescing on forks. There's a substantial amount of maintenance required to manage forks which isn't free: initial setup of repositories and build systems, reviewing, CI, and rebasing on upstream. Some of this is free-ish (i.e. hosting the repos on github) but other parts are limited/trial-size (i.e. CI actions, some AI review services) that require investment to scale. The idea is that people who are interested in maintaining a fork have their AI agents automatically share the ongoing work of review, CI, and rebase in a way where trust is established and reputation is tracked, in a way that makes it difficult for bad actors to inject malware into a fork.
I am currently on maternity leave with my first baby have just launched to the app store my first app!
I'm a product designer by trade and learnt to vibe code on Claude about a month ago now. I was weaning my baby and he developed a rash and haven't been able to work out what the pattern was causing it. So I created an app for weaning parents to log the food, add any reactions so when talking to their GP they can start to see a pattern and all advice, like nutrients, food variety, how to serve food all coming from the NHS and FSA.
Currently got 12 mums using the app and would love any feedback to make it even better! <3
To help my kids preparing for an exam (and steer them away from asking Gemini for solutions), I’ve built an iPad Kids Tutor app.
They can take a pic of a problem that they struggle with (initially for math but it works for any subject) and the harness then prepares a plan how to work through the issue using both claude and codex on the backend.
It’s been fun to build and watch my kids start using it and actually enjoy it (ok, not always).
A month and half ago launched Maxxmod, a Chrome extension that improves the YouTube experience by removing distractions (hiding Shorts and cleaning up search results) and adding new tools and shortcuts (volume booster, screenshot tool, quick access to Picture-in-Picture and playback speed controls). I’ve been adding features since then.
Last week, YouTube introduced GIFs in comments, so added an option to stop them from autoplaying and play them on demand (a play button).
Vela! A new desktop environment made to use just Vulkan 1.4, expecting to be faster than Plasma* and made with wlroots, but using a custom renderer. It has focus on fractional scaling, sharpness and usability. Take a look at https://velade.org
*currently building a benchmark suite, reaching approx. 40% less ram usage, 40% faster input lag, less cpu usage in some activities, and faster responsiveness during high cpu usage (from other tasks)
I found it bizarre when Microsoft moved the Windows start menu from the corner (Fitts' law, corners are infinite-size targets), and I see you've done the same. Why?
- finishing porting GNOME to Fil-C. It is starting to work but with some features still disabled.
- about 1/3 of the way through getting Rust to compile to Fil-ABI. This means you can interop between Rust and Fil-C without weakening either language’s guarantees. I’ll need this to get librsvg to work in Fil-GNOME.
- almost done making assembly memory safe. The SaRCAsm (safe runtime capability-enforcing assembler) works and I have OpenSSL ported to it but I don’t yet have 100% confidence in this. Once this is done I’ll be able to get rid of zunsafe (the closest thing to an escape hatch in Fil-C; it’s only used for the assembly in OpenSSL; without calls out to assembly, OpenSSL has too high risk of timing side channels).
I've been building a web browser in C from scratch, without any libraries and without AI.
This is finally approaching a usable state: navigates and renders Wikipedia, supports TLS, builds on Windows, macOS and Linux, loads PNGs and JPEGs, etc.
I'm going to opensource this once it supports most of CSS 2.2 layout and renders HN too (HN is totally blank right now, because it's a giant <table> which is still a TODO :-) )
I still don't know what to do with this. It's been a huge passion project and I'm very happy about it. I planned to create a book or video course about C programming, starting from printf("Hello world!\n") and ending up with a working browser; but the advent of AI coding agents in the past 2 years changed a lot the market for such a course.
I'm still finishing this due to stubbornness and some sunken cost fallacy effect :-)
Would love to hear feedback and ideas about such a project.
Knowing how to work without AI is valuable (what a surreal sentence to type). The first step of that is knowing it's possible. Years and years ago, when I got started programming, seeing projects like this taught me I can make anything I want, and I can always learn how to do it. There's no far-off wizard tower that must be relied upon to make everything for you.
What I'm saying is, I think this kind of thing will help the young'uns realize they can do it too. That fact is, unfortunately, far from obvious. One of the worst anecdotes I have from elementary school was of asking the librarian (who also oversaw the computer lab) who made OpenOffice, because I wanted to learn about how software works and figured that'd be a good start, and hearing her immediately dismiss the question by saying "People a lot smarter than us!" in front of the whole class. I can only hope nobody internalized that.
Also, out of curiosity from this topic being heavily on my brain lately, might I ask how you load images? Are you using libraries, or did you write any decoders yourself? Back in the day, my far less experienced self tried writing a BMP decoder, then learned about fuzzing and broke it instantly. Now that was a fun learning experience.
I recently did some playing around with having AIs test PNG decoders and it was as much of a mess as I expected. Broke a few in under an hour from casual prompting, and found weird bugs and oversights in every single one over a few days of not-so-casual prompting. (If you want to check reliability yourself, I found it eye-opening to run against a test corpus made from combining every major project's own corpus, and checking the pixel output, not just whether or not it crashed or ate all your RAM.)
Thanks for sharing the school story. I totally get behind that sentiment -- knowing how to build things yourself is very satisfying and you should believe that you can do it too. The earlier you get that lesson in life, the better!
I love how John Carmack put it: [1]
You can't know everything, but you should convince yourself that you can know anything.
The Handmade Hero series [2] channels this feeling really well, and I wanted to convey it again by building a browser from scratch but with a limited scope (Wikipedia, HN; no Javascript, but also no libraries and no AI).
I also loved tutorials in the 90s like "build your own OS", "build your own bootloader", "build your own C compiler", etc; build your own printf(), your own malloc, etc. Learned a lot this way!
might I ask how you load images? Are you using libraries, or did you write any decoders yourself?
No libraries means no libraries! :-) The decoders are written by hand, following the specs. They don't support every single configuration (e.g. JPEG has TONS of variants), but good enough for the media in Wikipedia.
This browser is a learning experience and it's definitely NOT secure (e.g. TLS "works" but it doesn't verify server certificates at all...). The decoders use a set of shared utilities for decoding big-endian/little-endian, 16 bits, 32 bits, etc and checking for overflows (using setjmp/longjmp for convenience). I think it's pretty good, and I also bet that Claude would find many problems. I'm prioritizing (1) code that works, then (2) code that's performant, and then (3) code that's clear.
I'm ambivalent about throwing AI at this codebase. It would definitely find issues and improve it, but I also want to keep the spirit of "Do It Yourself" and the feeling of "We built this and it works".
Now, when you say no libraries, exactly how far does that go? I unfortunately, a couple years ago, suffered through insisting I wanted to use C++ without the standard libraries, not even crt0, and finding documentation on how anything works was a huge pain in the ass.
Actually, what sold me on bothering with LLMs was asking one, last year, how to call a kernel syscall with GCC's inline assembly syntax, and seeing the answer be exactly what I needed... after I had already cobbled everything together myself! (I do not consider random mailing list archives to be adequate documentation!)
I didn't mean to say you should have an AI inspect your code, though. (I've personally found value in that technique, because current frontier models can recognize problems way better than their lackluster ability to avoid writing those same problems repeatedly into their own disgusting code. I can still just write the code myself, with another eye on it.)
I just really like fuzzing and really like differential analysis. For example, one of the images in my test suite decodes to one set of pixels with FFmpeg, libpng, stb_image, among others, and decodes to a different set of pixels with libspng, lodepng, Java's ImageIO, and other others (as far as I can tell, the reason is ambiguous 2nd-ed spec wording that isn't obviously ambiguous when you read it). You can quite easily fuzz one implementation against another, running them in parallel against the same inputs, and make a mismatch crash the test runner so the fuzzer flags an error.
Incidentally, asking Claude Fable 5 to do that usually triggers a cybersecurity block for me. Another point in favor of knowing things by hand...
Not that I'm saying you should drop everything and do this. I just like talking about it. Learning-experience projects get easily spoiled by trying too hard, which I did learn the hard way. Then again, I hella enjoyed practicing becoming aware of code coverage when I wrote a few prototype compilers. Researching online when I was making my first one is how I learned about Jonathan Blow and Casey Muratori, by the way. I don't follow Handmade Hero, but I do wholeheartedly approve of its existence.
I also loved tutorials in the 90s like "build your own OS", "build your own bootloader", "build your own C compiler", etc; build your own printf(), your own malloc, etc. Learned a lot this way!
You must be older than me! I was born in the 90s, and the usual reaction I expect to saying I built my own memory allocator is incredulity because "why make something that already exists?" or "ew, you should've used Rust".
I haven't run across that Carmack quote before, but I've told plenty of people the same thing in my own words. I think one of the most important parts of it that I've learned is this: You can't get something right without screwing it up first. I don't mean that making mistakes is a goal to target. I mean that you can't learn without the errors along the way. If you're always right the first time, that's just luck, and you haven't learned yet.
And I must say, I miss the days when "grok" was a weird word some programmers use, not an AI brand.
Anyway, I'd be quite curious to see your project if you ever put it up in public. I wouldn't worry about polishing it, really. In high school, I still didn't know people normally write rough drafts of stories and draw sketches long before making the finished piece. Never saw any of that, so I thought my inability to "oneshot" these things with the results of a high-skill artist meant inherent lack of talent. If all you ever see is finished refined products, how can you possibly know better?
Now, when you say no libraries, exactly how far does that go?
No libraries at all like stb_image, openssl, libcurl, etc. But it does use the platform APIs to create a window, get keyboard and mouse events, create TCP sockets, etc. It uses some libc calls like printf and malloc, but then it has its own memory allocators and arenas for DOM objects, CSS rules, and other common objects.
On Windows it just targets Win32.
On macOS it uses a bit of Objective-C to create an NSWindow and for the main message loop.
On Linux it uses libxcb to connect to X11. I planned on using Xlib originally, but libxcb is closer to the protocol and IMO cleaner. After using it I also got a sense that the X11 protocol is pretty straightforward, so I might just replace libxcb with a custom implementation.
For example, I need to query DBus for things like the desktop theme preference (dark/light) but didn't want to link build the DBus protocol messages from XML in the build, so it just queries libdbus directly and parses all the messaging. Maybe it should have a custom DBus client too...
(Linux doesn't have a "native" UI, so I made this exception on that platform to be able to make windows, etc as in the other platforms)
---
There are lots of reasons to "do it yourself" (and lots of good reasons NOT to do it too -- I wouldn't do this professionally!)
My goal here was to make a presentable and polished product, so I spent some time on code quality, correctness, performance, etc. and in presenting a nice browser that feels familiar. Here's a screenshot: https://imgur.com/a/ojWvciR
More than anything, I really like being in the flow and feeling the joy of programming -- building something step by step and seeing the results incrementally. A browser is quite a nice project for that, since you have the whole web to interact with and lots to implement.
> This is finally approaching a usable state: navigates and renders Wikipedia, supports TLS, builds on Windows, macOS and Linux, loads PNGs and JPEGs, etc.
So you’re implementing TLS (including RSA, elliptic curve, etc), JPEG decoding and PNG decoding from scratch? :)
TLS is always version 1.2, and only the mandatory cypher suites are supported: TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 and TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256.
This requires SHA256, AES and two elliptic curves (x25519 and secp256r1).
AES is used in GCM mode, so that's implemented too.
However, the browser doesn't verify the server signature over its certificate. So ECDSA and RSA are not implemented :-)
Getting TLS to work was indeed one of the more frustrating parts. The cryptographic functions are very well documented and there are lots of good test vectors, so that wasn't too hard (just a lot of work). But getting TLS working was very frustrating -- the handshake would complete, I'd switch to encrypting mode, and... the server would just disconnect my client instantly, with no further feedback or a stacktrace or anything.
I ended up debugging that by building openssl with LOTS of printfs all over the place and fixed nonces, talking to a local HTTPS server, and tried to make my client emit exactly the same bytes until I found the issue -- memory corruption, of course :-)
I built on amazing tutorials available online. For example, elliptic curve cryptography and x25519 in particular are SUPER well explained at [1].
JPEG was fun too. Tutorials like [2] gave a great overview of how it works, and the spec [3] is super clear (but includes TONS of optional stuff that can be skipped)
PNG was the easiest, the RFC is very straightforward [4].
(I actually did the PNG decoder before doing fonts, because the TTF format is so crazy that I always decided to using bitmaps fonts loaded via PNGs instead)
It's super satisfying to build ~300 sources in ~2 seconds from scratch using just MSVC, Xcode or Make, and then launching Wikipedia -- no libraries, and it works :-)
Let me also list some of the big things that are NOT included:
No Javascript :-) React sites are completely blank, like instagram.com.
No GL/DirectX/Vulkan. It's all software rendered and it just blits a pixel array to screen on every frame. Highly portable though!
Font rendering is antialiased but it's not hinted, and small glyphs at low DPI look start looking blurry.
No Wayland support, only X11 because it's the most widely available. I might do a Wayland port later.
All in one process, one thread per tab. No process isolation (also something to do later)
Only HTTP/1.1 and TLS 1.2, no HTTP2 or later. No IPv6, only IPv4.
No SVG, no <canvas>, no form controls (buttons, etc.), no cookies support (for now...)
The CSS support however is bigger than CSS 2.2: there's var(), --custom-properties, @media queries, etc. and I'm planning to do flex and grid layout too (because Wikipedia uses that a lot).
It's very encouraging to see that there was a positive reception on HN, this is the first time that I talk about it publicly. Really enjoying the extra motivation to getting this released now :-)
The first commit of v1 was in Oct 2022, so it's been about four years.
Just looked at a few more metrics:
v1 was 148 commits over about a year; gave up when it came to TTF font parsing and rendering.
v2 was 1075 commits over 3 months and in C++, gave up after doing sockets but realizing that TLS and crypto was going to be required. I also strongly disliked the C++ codebase :-)
The current repo is v3 and back in C, started in Oct 2024 (2 years), and it's 4822 commits. About 600 .c and .h files, about 68k lines of code (14k blank, 7k comments).
I had a lot of previous experience with networking and systems programming, but not with rendering.
I haven't been building in the open, because I wasn't sure I would ever finish this! :-)
I'm currently planning to opensource early next year and I will definitely post to HN.
(The original name for this project was "Hacker's Browser", as in, an HN browser :-) But then I figured that Wikipedia worked pretty well and grew in scope from there...)
I'll also be looking for early testers soon, to verify that this thing builds on various platforms before the public release. If you're open to try this early on then see my profile to get in contact.
Working on http://downread.com/, making it an advanced RSS reader.
I am trying to push the boundaries of what an RSS reader can do. Just added podcast capabilities, complete with background play, queuing and video streaming. Now I am working on adding posting and commenting within the reader, based on rss.chat and the webmention protocol. I am also procrastinating the completion of the mobile apps, which I really should finish soon..
I got a 360 camera (GoPro MAX v1) for my bicycle, after re-discovering bicycling during COVID. I soon found it was fun to review the video after a ride, but the software was a bit clunky. Upgraded to a DJI Osmo 360, and found the software likewise powerful, but still not what I wanted.
As an audio video developer I knew I could make it work the way I wanted, and this year I finally set about doing that: https://recap360.app (macOS-only at the moment).
It’s a ~5MB download from the macOS app store; no in-app ads, no accounts, no ~GB download. It loads the video of a 5-hour bike ride in seconds, direct from MicroSD. Complete with a map overlay & graphs of various internal metadata. (Speed, altitude, ascent rate, but also internal temperature etc).
It’s been fascinating to see the various tradeoffs the camera vendors make; how much in-camera stabilization & rolling shutter compensation vs how much to leave to the player. Also interest ing is how the metadata is recorded in the file; some document them, most don’t.
Started with GoPro MAX, added DJI Osmo 360 & Insta360 support. More coming soon. Even plays dash cams & body cams (Vantrue & Boblov; I was curious to see how they stored GPS data).
My wife and I are working on a math/science/CS-inspired jewelry: pieces that are beautiful but have a (maybe a little hidden) meaning. E.g. earrings that are actually the y-combinator from lambda calculus.
We have two styles as of now: lambda calculus based pieces (the y-combinator, and an expression that computes Graham's number), and quantum computing circuits.
We had our pieces in a jewelry store, but unfortunately the owner retired.
We are trying to do pieces that you can "read" if you know how. E.g. the quantum computing circuit piece is an circuit to compute a non-local catalytic control-not-not gate (and we have lengthy science explanations on the site) -- soon to be in an article on arXiv!
So we would not just incorporate golden ratios. But if there is a nice way to represent a specific elliptic curve that would be nice. I would appreciate if you have specific pointers.
The combination of beautiful, true to the science, manufacturable, and wearable is hard to get right. And selling is even harder.
I'm also writing a puzzle game with a door that has one million locks. It's real-time multiplayer; some locks have opening combos, and others can be opened only by > 2 players. I'm writing it to practice with PostgreSQL. I only have an HTML prototype to test it. https://6251.s3.eu-west-1.amazonaws.com/hn-door-screenshot.p.... If you are interested, I can publish the code.
I'm building a free/open source modular audio composition environment, called LoomDSP. It's inspired by Eurorack modular, VCV Rack, and Bitwig Studio's The Grid. Please feel free to suggest ideas and report bugs!
I live in San Francisco and we get some beautiful sunsets here! Over the past few weeks, I cobbled together a system of alerts for when the sunset is predicted [1] to be particularly outstanding, resulting in this page that displays the prediction and also shows the current view from several cameras around the city.
[1]: Predictions are LLM-powered largely to more reliably consume the free-text descriptions in traditional (e.g., NWS) weather forecasts of predicted high and low cloud cover around sunset.
I'm trying to build "the most accurate on-device transcriber". I'm doing a custom train of Parakeet Ultra that fixes the model families' issues with digits by having the model output them in spoken form. I'm also trying to get the harness it sits in to better merge the 15s windows. https://mimicscribe.app/
I'd like to get into ML work, so this is a showpiece if nothing else. It's still a little broken.
Because I can. Well, let's be honest: because LLMs can. There's nothing particularly novel about this. It's just wiring a bunch of libraries together. Certainly would have been time consuming, but not difficult, strictly speaking, since all the heavy lifting is done by the libraries utilised.
Likely going to migrate https://breaka.club HUDs and editor UI to React since Godot control nodes are by no means impossible to work with, but somewhat cumbersome.
EDIT: Oh, will be open source like a lot of Breaka Club's other libraries very soon. Still optimising.
Implementation is predominantly Rust, with TypeScript. I don't use GDScript at all.
I'm the current maintainer of GodotJS, although not its original creator. My games are built in TypeScript, and I use GDExtensions (typically C++) when something is performance critical, like the rendering for this project.
Definitely slower than I'd like, but it's probably worth noting there's a few things that make this demo much slower than it needs to be:
1. There's some stuff simply not well optimized internally. Working on these now e.g. there's unnecessary internal hierarchy on all nodes that only needs to exist on a small subset (related to border boxes, overflow etc.)
2. Eager shader compilation. There's some glass and blur shaders that are compiled on load. These really shouldn't be.
3. Godot 4.x on the web is always pretty slow to load. Time of first contentful paint (if you ignore loading screens) is nasty across the board. Godot 4.8 upgraded Emscripten, so there's actually potential to improve this quite a bit since it's now easier to dynamically load WASM (which is what we're doing extensively here). I'm unaware of any work on this upstream though.
4. THE BIG ONE. This demo is a bit of a monolithic beast. It doesn't just include Godot's implementation of React. You can reveal the code for any page in the demo, and edit the code live with an inline text editor. Genuinely bundled in this demo is a full TypeScript 7 implementation (based on Bun's Rust implementation) that type checks and compiles code to JS on the fly, and hot swaps components on the page live.
For the most part, this isn't designed as a performance demonstration, although there is a small section on repaints/relayouts. It's mostly just a demonstration of what's possible with the library.
Realistically, there's also the fact this is React. Its architecture is based on React Fabric (like React Native), so runtime performance should be pretty decent. But frankly, it's never going to be as performant as some alternatives — it's not really the primary goal of the project. I'm considering an alternate SolidJS front-end for improved runtime performance, but that'll do nothing for first load time, unfortunately.
I'm trying to build a way to manage all my open tabs in a sane way! So you close all the tabs and it automatically saves them all to a to-read list. It just got approved on the chrome web store, and I'd love your thoughts
"Gave" is Norwegian for "gift". It started as something I built for my family as each Christmas there were numerous bad solutions to making sure nobody knew what they got while still making sure that nobody got double up of something they wished for. I have iterated a lot over the last 3 years: I've built one version as a very simple web page, one version as a PWA, and now a fully-fledged native iOS and iPadOS app.
In Gave, the people who follow you can see what's free, reserved or bought, so nobody receives the same gift from two different people. You never see the status of your own wishes as the server never sends you that data. That goes for notifications, widgets, Siri and Spotlight too. There's a "Hide status" switch for when you show your screen to the person you're shopping for.
Some other things it does:
- Private notes and secret gift ideas.
- Easy view of whose birthdays and holidays are coming up, and if you are prepared.
- MCP server for connecting your favorite agent.
- Very native experience: It is integrated with Spotlight, Siri, deep linking, notifications and Apple Intelligence if you want to use it.
It's free: no ads, no subscription, no analytics, no affiliate links. Data lives on a server in Norway. Native SwiftUI on iOS 26, Laravel and Postgres on the backend. It now has English support as well, but the default language is Norwegian until the next version goes through review (hopefully today).
I'd love feedback, especially from anyone whose family has tried a shared spreadsheet or another wish list app and given up on it.
Currently researching Create ML to try and specialise Apple's foundation models
to some particular task(s). Also adding new features to some indie apps I published:
I'm working on a Bluesky client that uses on-device AI to organize and summarize the contents of your feeds, so you can find the interesting posts faster. You can choose among dozens of models. This is the area of the product I'm experimenting with most right now. Having a few more people provide feedback would help me.
In my opinion, the overall look and feel, especially regarding usable screen space on small screen devices, is superior to the default client.
My personal page with contact information is linked in my profile here. If you want to give it a try, just ask. Please put "butterflyz" in the subject line so I can keep track of inquiries. Thanks!
A minimal X11 application that prints state and keycode on key press. I needed the state for special function keys like Alt-GR, but I didn't want to search the web for it.
Originally started from the st[0] terminal and removed unused functionality, but there was still too much code. Ultimately I followed a minimal example from rosetta code[1].
For the task of removing functionality I developed a small tool based on LLVM that prints the source code location of functions that are unreachable from main, so that I can safely remove them without altering functionality, deleted ~3000 LOC in a couple of minutes[2], still I'm struggling with integrating the tool with cmake
Very different than usual HN vibes here, going for some solarpunk-y & creative coding projects:
https://cybersolar.earth - artist website for my solarpunk hiphop & drum n bass album coming out, plus some thoughts on climate+culture
https://newsdash.concourse.codes - basically: Perplexity for climate news. hardcoded to be personally relevant rn, uses Gemini to do some basic web search grounding. Turned out to be fairly helpful
https://arena.concourse.codes - Use are.na channels to 'fine tune' a Gemini image gen model and prompt what you want. Inspired by wishing I could easily tell Midjourney to quickly use a collection of 10 images to set the style of an image. gens limited as is WIP but fun to play with for a min (edit: ofc i share today & have UI bug! click the page header to fix for now, will fix later:D)
Working on Zoupyu (https://zoupyu.com). You paste a long video, it suggests the best moments, and you pick the start and end points which then become clips. Most AI clippers just pick for you, so I wanted something that gives me full control over each second. I want to expand it into a platform that brings creators and clippers together through content rewards.
7 months ago I left my job to give more time to Zoupyu. Solo building and figuring out everything from SEO, marketing, tech, design, and everything else it takes to build and run an online product. I never sold anything in my life before and now we have around 50 paying customers from all around the world, just 3 months ago my goal in life was to make one sale just one! and now its around 50. I know its not much but feels so good that I achived that. That too all organic.
I started working on a tool to help us deal with the increased output from AI. Instead of reviewing large PRs, we review architecture diagrams with proofs. Most of the companies that my friends and I work at had stopped reviewing code, because it just became unmanageable. Also, working a 9 to 5, reading large outputs from LLMs got exhausting, even with the skills/plugins you throw in. The problem went beyond decision fatigue, I felt like I had a hard time understanding what the agent wanted to do or what it had just completed.
So I built https://draw.redcrayon.ai (with a few friends), inspired by tools we love like Excalidraw: a live architecture board you and your coding agent draw on together. The agent connects over MCP (Claude Code, Cursor, Codex, claude.ai) and draws what it changes as it works. The edges are meant to cite a source line (or fingerprint). You read the board instead of walls of text, and you can circle something in red crayon to get the agent to answer/fix it.
What we use it for:
- An incident scratch pad, where the agent lays out what it finds while debugging
- Seeing how a change affects the architecture before reading the diff
- Flagging the risky parts, like a DB migration, with every piece of code that reads that table
- Github App that generates a diagram per PR (something that is a work in progress)
- Ever wanted to create a branch of your existing diagram rather than completely clone/copy it? That's something we added to draw.
I building a service to let different AI agents talk to each other. You can use it to connect your own agents, e.g. Muse and Grok Bot, or make Claude review Codex’s work. Or you can use it to connect to your friend’s or colleague’s agents - think Grok Bot team but without the requirements that everyone has to be part of the same org.
Why not use mail? Well, for one, it’s hard to register many different emails programmatically, and you do want a separate inbox for each of your agents. And you shouldn’t really give your agents access to your primary email: it’s not secure, and you will be drowned in noise past a certain email volume.
There is a lot of competition in this space already; it’s heating up. Wish me luck!
A new editor and processing engine for the Foveon Sensor cameras that were manufactured around 2002-2019. The existing software is either wildly slow, or does not process according to manufacturer intention. Think choosing between waiting multiple seconds after every single UI interaction to display changes, or color casts not being handled and the images looking green/magenta in different areas of the shot. One tricky bit is that the 4 different sensor architectures are all processed differently.
This weekend I am modifying the pipelines to be able to operate on downsampled sensor layers so that I can get multi shot editing as snappy as possible. If you are only displaying 500ish pixels wide, its wild to still be rendering the full 16mp image. Perfetto has been very helpful while optimizing.
Up next is redoing the caching/scheduling architecture to reduce the amount of overhead and to support multiplexing downsampled editing so that the images are processed within the same GPU pass instead of N passes * M images serially.
The idea is that you enter in the profiles you want to follow across Instagram, X, Facebook, Youtube, Linkedin or TikTok, and Narro pulls the posts into one place without ads or the algorithm.
I've been having fun with using AI to gather data on niche subjects so I can visualize and understand it better.
Recently, this has been porting the wikipedia "Getting to philosophy"[0] to other gaming wikis, and working alongside agents got me to a really nice visualization[1] for it.
I'm now slowly working towards converting this to an immersive article so I can share some of the emerging patterns and evaluating the differences within the various wikis.
A suite of high-dopamine games which use smart exercise bikes as the control interface.
When playing Zwift, Rouvy, etc, I find that I get too competitive and ruin longer workouts by trying to keep up with or sprint past people who are much faster than me over long distances.
This is bad for building stamina and endurance, but I get bored, and it's hard to fight boredom and physical strain both at once.
I do not get bored sitting in front of Balatro, Raccoin, Tetris - anything with bright colours and satisfying noises and combos.
The first finished prototype game is very simple: you sort coloured marbles. Higher watt output from the bike means marbles are blown, by wind, upwards as they roll across the screen. Lower watt output means they fall. Steady watt output means they move horizontally. Marble goes into correct hole -> satisfying noise. It rewards steady watt output over time, with minor adjustments, which is what I'm trying to work on right now.
I was pretty amazed by how engaging I found it (total iPad baby brain). The miles really fly by - I get tired before I get bored.
I think that next I'll either do a tower defence game, or a less-dopamine-rewarding game in the style of Kerbal Space Program. Taking your actual weight and watt output, putting a bike with wings into a physics engine, and the gameplay loop being unlocking increasingly large airplane pieces, and designing - then flying - increasingly ridiculous airborne bicycles seems like a compelling idea, though less good for fitness.
The control surface being so limited - essentially one measurement which can either go up or down - is a really fun constraint for creativity.
Very cool! I was actually thinking about going the exact opposite direction.
There’s an open source “recomp” of Gran Turismo 3 that supposedly works pretty well on a laptop. I kinda want to hook it up so that the Zwift ride provides the power output for the car.
If 1w = 1hp then I could probably win some of the first career races right now. And if I get back in shape, I could eventually hang with the base model WRX’s.
There's something in this idea. you can actually tie it to a smart watch (Amazfit allow for custom apps trivially for such uses), and you have a bigger pool.
Thanks, yes, I agree - it really doesn't "feel like" exercise when you're being bombarded by dopamine rewards. Despite this, most "exercise games" leave me with the same feeling of boredom / "fake fun" I got from playing educational games on the computers in the IT room at school.
I'd love to try a similar approach to gamifying exercise on a smart watch. That's an even more fun problem, because (when running, for example) you don't even have a constantly-present screen to communicate dopamine to the player with. Perhaps some watches wouldn't even be able to provide sound feedback? I've never owned one. It's a really interesting constraint for game design.
I’ve been working since the start of the year on a minimalist web reader (https://papercove.app) that I use every day to follow RSS feeds, newsletters and read-later, and that helps keep things from piling up.
I thought I’d get where I wanted much faster, but it turns out I ran into quite a few challenges: it’s quick to build something simple, but it takes much longer to get a reader that’s robust and pleasant to use (extraction, parsing, security, infrastructure, etc.).
On the tech side, I’m using my usual stack, Laravel and Livewire, with some Python for extraction and Go for the email infrastructure. This month I’m working on the beta release, since several people around me were already using it, so I’d like to gather more feedback.
That is a great landing page! I can relate on the velocity side, it can be quick to build a feature, but it takes time to get everything feeling right and usable.
A few months ago I was looking to create a dead simple weather site that would show me the next 24 hours in plain language and email me the night before. Most weather sites are horrible to read and most of this information is freely available by your local NWS.
I used to really enjoy watching Dota 2 tournaments. I also like Bake-Off, and Battle-Bots was a great idea.
I'm tinkering on a project to take two contestants and dump them into a reasonably nicely equipped linux environment (checkpointed fly.io Sprites), giving them a task, and watching them compete in real time to see who gets it done first. All of it streams to twitch.
Here's an example match (I have two agents pretending to be contestants for the sake of the demo: https://www.twitch.tv/videos/2923688013). Basically anything goes; I imagine people will clone+stow their dotfiles, leverage their workflows/harnesses/editor configs, etc. Viewers will get to see how its done.
Next on the todo list is "weight classes" that let contestants compete with a fixed level of harness capability: natty means no agents at all, small local open weight LLM tier only (qwen, gemma, etc), large open weight models only (deepseek, glm, kimi), frontier lite (haiku, luna), frontier (6.1-sol, haiku), frontier plus (astra, fable).
Ultimately I want to rank users in each weight class so people can compete for delivering the most at various cost tiers.
I've been burning the midnight oil and plenty of tokens on two projects.
Project #1:
The first project is MiSTerVision, a CRT-friendly Jellyfin and Plex client for MiSTer FPGA built in Go. I have a MiSTer FPGA connected to a late model Sony Trinitron CRT and wanted to watch movies and shows I've ripped to my Plex server.
I originally contributed PRs to the MiSTerFin project, a Jellyfin client for MiSTer, but the project was stagnant so I forked it, ported it to Go, and was off to the races. I added Plex support, improved the UX, improved performance, filled in a bunch of compatibility gaps in the Jellyfin support, and released it for use by anyone with a MiSTer.
Watching old-school cartoons, like Dungeons and Dragons (my favorite) on a 4x3 CRT in better fidelity than I ever had as a kid is sublime. It's had 25 or so public releases over the last few months.
Project #2:
The second project is HyperServer. HyperServer is a Go framework for server-rendered apps with HTMX that I've been working on off and on for a few years. I was originally inspired by Pagoda and really like the architecture and ambition of Pagoda but didn't like its dependence on third-party libraries.
HyperServer uses standard net/http handlers and html/template, requires no ORM, and makes authentication, sessions, and storage optional. It’s intended as a reusable library rather than a starter repository you copy.
I haven't been confident enough to make a "release" for it yet, so it's labeled as "experimental". But I have built a functional (and fast) blog on it. I plan to use it as a platform for a couple of other web projects I'm scheming. Hopefully that will prove its utility and I'll make an "official" release.
That said, anyone interested can peruse it and use it as-is. Any feedback would be welcome.
As part of my previous job I tried to deploy a local model on a client's environment, then I learned about the balancing act called inference optimisation, is was a huge headache back then and it gave me the idea to create Tesnward - a local AI inference optimisation tool - Tensward detect your environment and help you optimise inference, it runs completely local
https://github.com/amterp/muster is a Libghostty-powered terminal for managing your agents. I just gave it a bit of a visual refresh this weekend and it's looking pretty great imo. It's my daily driver lets me keep an eye on my agents and allow them (cross harness included) to work together, communicate, and spawn new Muster agents themselves.
https://github.com/amterp/kan is an offline Kanban-as-files-in-your-Git-repo project I started near the beginning of the year and continue to evolve and work on. It's my main way to track tasks and projects, and these days I'm handing a ton of control of my kanban boards over to AI agents that interact with it thru the kan CLI, express which cards block what, etc, and they can pick up work automatically with 'kan ready'.
I've recently been making improvements to Larm, https://larm.dev, an uptime monitoring service I've been running for a while now. It's got global probe coverage, automatic/standalone status pages, and lots of alert channels. Pricing is simple and the limits are generous.
It's accumulated a decent number of users now which is exciting.
Under the hood it's an Elixir backend with a network of little go probes running all over the world, all connected by TCP tunnels. All the check result data is stored in Clickhouse. Status pages are generated on demand and served through a CDN.
AI-generated code comments or Markdown usually isn't my style, so I built an AI-native linter and formatter, which automatically fixes the style or asks AI via a hook to regenerate if something is misaligned but cannot be reformatted deterministically.
I’m learning English and wanted to watch movies in the original with my girlfriend. She isn’t learning English and didn’t want to watch in this way. I couldn’t find anything that really worked, so I vibe-coded this player. Now I listen to the original in my headphones and she listens to the dub on the speakers.
This player runs in a browser and lets you play two audio tracks from the same movie on different devices.
It doesn't require installation or an account and the movie doesn't leave your computer.
This is a lovely idea. I'm also learning English and have the same problem at home.
Nice that the movie never leaves the computer, too. How do you keep the two devices
in sync – do you correct drift during playback, or only at the start?
I'm building an open-source "spatial workspace for directing AI coding agents like an engineering director".
I got overwhelmed by walls of text of parallel Claude Code sessions. This limited the sessions I could steer in parallel to maybe 5 or 6. I knew there had to be a better way, and I found one: A canvas per project. Work laid out as cards. An agent in a worktree per card. Chat condensed to WhatsApp-style back and forth, not a litany of tool calls. Result is usually a narrated demo video that I can comment on by voice to have the agent go another round. Remote-control from my phone. Remote workers to scale beyond my machine. Everyone I know who tried it has stuck with it so far.
I've been working on a simpler, more extensible Markdown variant I'm calling Zd. I've also been working on a simple, embeddable language that's something in between Lua, Koka, and Erlang.
In both of these projects, I'm trying my best to find a balance on AI usage. So far, they both have held strongly to the principle of only accepting code I've written personally. At the same time, as I begin to land on stable design decisions I've found that speculative experiments with AI can be helpful. I can benchmark potential future decisions prior to committing.
I'm not saying that AI couldn't be used to do more, but I'm trying to ride a line where things still feel personal, local, and deeply my own design. And more than that, I'm practicing holding on to the joy in the work. It's too easy to just know that AI could be used to plow through a problem. It makes my scope tend to creep and my estimates shrink. So, pushing back involves continuing to scope small changes and remembering to feel positive about each win.
thanks for this. I figured it was always for starting programmers etc. but tried out the k8s course just now (ive been using it at my job for a year or so but mostly through vibes and memorized commands rather than knowing it with any depth), seems worthwhile, learned a thing or two.
Having AI/digital overload, I purchased some hand planes and am flatting a hardwood slab (and dimensional) that I had milled from a tree felled in my yard in order to make some furniture.
I've been working on G-Scope, a free application for digital microscopes, available for Windows, macOS, Linux and Android.
Most USB microscopes come with poor software, often limited to one brand or operating system. G-Scope works with almost any UVC microscope and several Wi-Fi models too, it has photo/video capture, calibrated measurements, annotations, and more advanced features like panorama stitching and PDF reports. Some pro features require a one-time purchase.
Everything is local, without account, no subscription, cloud or telemetry.
I’ve been building a little Python tool to extract a MIDI drum track from an mp3 or other song file. Eventually I want to buy an e-drum kit and learn to play along with my favorite songs.
From what I can tell, we’re about even with the current state of the art in academic publications right now, at least on MDB. Runs faster than real time in pytorch on a few years old MacBook.
It even does a decent job creating easy/medium/hard versions from the original “expert” track. I compared ours to some charts from Rhythmverse and while we’re not as good as the official ones, we’re pretty darn close.
Spent some time trying to extract an instrument model for the drum sounds from the recording too, but I think that simply requires too much data for lots of the songs that I like where the mix is really dense with a ton of sound from distorted guitars and cymbals etc. Works a treat on super clean mixes like Back In Black though.
Next step is probably to create my own rhythm game, since my initial experience with Clone Hero and YARG left a lot to be desired.
That sounds cool. For my uni project 10 years ago I chose to use neural networks to detect guitar notes, but not by pitch, but to detect exactly which string/fret was played (because you can play same note on different strings). It worked okish with a very basic network, quite suprised. I made a small game you could play with your guitar.
Since then I wanted to develop further such a game, and now I'm almost ready to release https://ultimidi.com , check it out, it's not really a rhythm game, but my goal is to add more mini-games to it (including rhythm games) and more instruments (now has piano, but planning to add guitar, ukulele too, not sure if drums would work too).
Very nice! Extracting the guitar track would be awesome but it’s so much hardier than drums. Even cymbals are hard. And guitars have chords and distortion and…
Yeah, that's why my idea was to use neural nets trained on manually labeled data, instead of relying solely on pitch detection.
Also, probably AIs nowadays can help a lot with generating training data, and even improving detection.
That being said, even SOTA voice recognition has only like 95% accuracy, so it gets wrong 1 in 20 words. Imagine if transcribing a melody and 1 in 20 notes are wrong, for a fast melody that would mean a wrong note every few bars.
I've been building AltStack.jp [1], a directory of digital services (cloud hosting, domain registrars, email providers, etc.) that are operated in Japan by Japanese companies.
The site is static, built with Astro. When it came time to host it, I didn't like any of the Japanese options. I could have put it on Cloudflare, but hosting a directory of Japanese alternatives on an American CDN felt a bit off. So I started building my own static hosting service: Kamakiri Pages [2]. It's CLI-first.
Note: AltStack.jp isn't hosted on Kamakiri Pages yet. I plan to switch it over within the next two weeks.
Feedback welcome on both, especially if you know of services that should be listed.
I always wanted to have all laws to be easy to interpret or at least try to track their changes over time to better understand what’s going on.
So I’ve built OpenNorme for the Italian legal corpus here: a single git repo with all laws from 1861 to today with daily scraping from an official api.
https://github.com/dej611/italian-laws-history
After every update the github pages is rebuilt with the new data, so you can see some trends in there with some aggs and nice charts. Also, had some fun to build a clippy clone with an in browser SLM.
Over the last year, I've been working on Review Box https://review-box.rstankov.com/ on the side. It is a tool to help me with GitHub Pull Requests.
It has an integrated dashboard for what to review, integrated symbol understanding, a clear split of reviewed/unreviewed files, and a result diff view (which shows me only how files are going to look in the end).
A proportional width font terminal. "i" and "w" shouldn't take up the same amount of horizontal space. We used to have to do that because computers were slow and we counted cycles and every bits mattered. We live in a land of plenty now though so we can spend cycles doing math so that ls -l/top/etc table output is still aligned but "i" and "w" can take up the appropriate amount of horizontal space.
The glue between AI assistants, agents, and data. 100% open source (MIT).
Provides a flexible data layer plus users, groups, granular permissions, notifications, durable work queues, audit logs and more… MCP and CLI interfaces. Works across all agents and assistants (Claude, GPT, etc)
Most useful for small teams and agencies that need to build one off solutions for collaboration and operations inside of the harness they already use.
Teaching myself how to build a Spiking Graph Neural Network in Zig, to run on my Brainchip Akida AKD1500 neuromorphic accelerators (and Efinix Titanium Ti60 FPGAs).
Good times, but lots of foundational math I hadn't previously been exposed to.
Some reading material if anyone else is interested in that:
I continue working on my side project Babel Nexus. I had some issues with billing so I moved it to my home computer and am rebuilding it using powershell and VS Code. I am trying to figure which small AI system I should put on my computer like Cline, Zoo Code, or Qwen. I can do this because I have an NVIDIA graphics card. I have also looked at Ollama. I am steadily rebuilding my app. It costs very little to build it on my home computer. I have redone a lot of little things and added a whole bunch of content. I plan on using Railway to host it because for the long term it is less expensive. Right now, my existing app is placeholder. https://babelnexus.com I added several new free ebook applications recently to the Beyond the Public Domain section.
I've hated all budgeting apps I've used and found them to be always restrictive, expensive, or envelope (YNAB) focused. I don't really need a "budget" app, more of a personal finance app. I just want to be able to review all my family's transactions and look at cash flow and track net worth, so I built my own. A few of my friends were interested, so I gave them access. I'm seeing more and more of these budget apps bring in LLMs which is such a bad idea. The only AI I built is opt-in and uses Jev AI to auto categorize transactions that are stripped of info.
A: yes, transactions are tied to each individual account. the transactions pages brings them all, and then you can filter by anything like the actual account, merchant, category, tag, owner associated with the transactions, etc.
I was laid off recently, so I needed to update my knowledge to be able to pass interviews. But I’m a slow learner, and I needed to accelerate the process. So I built Tutor, an open-source local project that extracts all the concepts (using AI) from study materials (I started with books and online documentation, but recently added YouTube courses as well), creates lessons in simple language, and also tests you. It helped me tremendously to get out of my brain-freeze state.
I use it with GPT Luna and Haiku (project being local, you add whatever API provider you prefer, with your own key), and both proven smart enough and dirt cheap.
I made a website mostly to post coconut water reviews. I never really liked coconut water until I tried an unpasteurized option and that led me to spiral down the rabbit hole of coconut water terroir. As far as my taste buds are concerned, the best coconut water is coming out of young nam hom (fragrant water) coconuts from central Thailand.
I'm not really interested in making any money off of this site at any point but it has been fun to see the Google analytics data as a few dozen people visit it every week to read reviews. I've reviewed like 30 different brands at this point.
Building a development environment to run coding agents in the cloud. There's a lot of other products in this category. What makes ours really special IMO, is the built-in task management system. Your agent can split up large jobs into subtasks itself and run each one on a different agent / model / machine. You can clearly see where things are at right now. In my personal usage, I can run tasks autonomously for weeks at a time this way and just provide steering input to the coordinator.
I have developed an app designed to encourage in-person meetups among groups of friends. It makes it easier to organize outings for people who already know each other. You place a pin on a map and share it with your group. Those who already have the app receive a notification. Those who don't can be invited through a universal link. Even without the app they can see the location, the time and the number of confirmed attendees on the web (the app is required to confirm attendance). I am building it on my own with FlutterFlow, Firebase and local SQLite, in five languages.
This month, besides promoting and improving the app, I've had to spend time defending it: bots used the SMS login to trigger 1,744 code requests in nine days. Only three of them ended in a sign-in. I have since added several defenses, including a daily check that compares codes sent and sign-ins per country. None of this fixes the app's core problem. Open the map before any of your friends use the app and you find it almost empty.
Still working on https://github.com/Disla-Novo/Dimidium_Bellerophon, a DSL for 3D printer automation.
I’ve been putting off the networking and printer management side, but I had an idea I’m excited about which is a non-planar slicer that outputs my DSL instead of static G-Code, which could allow more dynamic, point-by-point control. Right now I’m mapping out the features and how it would work. It’s currently an IDE I built on Java (Spark). I’m considering moving the backend to C# but haven’t figured out where to start.
I've been working on a self-hosted cloud drive (like a Google Drive alternative to run on your own server), about to release v2.
v1 didn't have a desktop app with sync, so I added one. I also redesigned the whole web app (a few pages are same as old design but I might change those later)
Honestly, I rewrote almost everything. I've been using it myself since the first alpha, and the v2 release candidates feel like a different app.
It still has some rough edges, so I got a friend to run it on his server as a tester which made finding bugs a lot easier.
My partner and I are both software devs. She quit her job a year ago, and has been working on a few micro-businesses to try a) learn some new skills and b) diversify our sources of income. So far this year she has built:
As a heavy LLM user in my daily work, I've been working on optimizing agent memory on complex projects. That resulted in this simple Rust-based tool called Wayfinder: https://github.com/zaterka/wayfinder
I've been trying to improve agent knowledge over a codebase, and existing approaches weren't giving the result I wanted, and I wanted to keep it simple. In the end, I added a semantic layer on top of an LLM Wiki and it saved 30% of token usage in multiple benchmarks I ran. It also adds some nice features to deal with concurring agents updating the shared memory, to ensure all information there is reliable and fresh.
It's published to PyPi, so an "uv tool install wayfinder-memory" already does all the work, and it's compatible with almost all Harnesses.
This is super cool, I've been thinking in this direction as well.
The codebase memory knowledge graph should be maintained alongside the codebase whenever possible, which basically means checking the graph into the repo alongside the code, or some other architecture that has an equivalent effect. But then it complicates the design of the retrieval indexing; I think your SQLite rebuild idea might work for that though.
I have a question for your design: since your knowledge graph is checked into Git and therefore Git worktrees should already isolate them and resolve conflicts between branches, why did you include the section about concurrent agents as well? Is that for maintaining the SQLite index which is not checked in?
A self hosting dashboard that gives you a Railway-like canvas to connect your different services in a project and that uses tailscale for easy multi-server setups
I'm doing something ambitious. I'm taking a radical approach to improving rust compile times. The reason it is radical is it involves building a brand new rust compiler. This was inspired by a project I started[1] about six months ago to teach agents to build C++ compilers from scratch. I wrote a (long) series of blogs about the process[2].
After iterating on many agent built C++ compilers, I now have a version that compiles about 4x faster than GCC and Clang[3]. I think i can extend the same approach to rust. I'm still a ways from the rust self-hosting stage but early results are promising.
My current Rust program compiles in 25 minutes. So I'm all for this. I believe the world is gravitating more and more towards this programming language. I've read the other day that Discord is rewriting its entire program in Rust for example. A lot of people would benefit from reduced compilation time.
I'm revisiting the Netflix Prize challenge with LLM. So far I've hit the 9% improvement mark within a couple weeks. I seem to have hit a wall in the last week or so with no further progress. I'm hopeful to hit the 10% goal with one of the many experimentations though.
I'm exploring using XMPP as a coordination layer for multiple LLM agents.
I'm using Herdr to run multiple agents in parallel, and my goal is to see whether I can get them to collaborate more efficiently.
For example, when one agent's PR gets merged, other agents immediately become aware of it, rebase their own PRs and fix conflicts.
What also often happens is that new tangential requirements and sidequests often arise while working in a session, and I now let agents hand these off to parallel Herdr tabs. The handoff can be done with or without XMPP. These parallel agents announce their statuses in the XMPP MUC (groupchat), which enables them to coordinate.
Technical details:
Converse XMPP client runs as a headless NodeJS daemon with a CLI that can be called by agents.
I created a /coordination skill which is enabled by default via a Claude Code hook. When an agent starts, it ensures the Converse daemon is running and then joins a password-restricted and members-only MUC where all the other agents in that Herdr workspace are as well (and also myself). I can then assign that agent tasks via messages in XMPP (DMs or groupchat) or from Claude Code.
Once the agent has the task, it announces it in the MUC so that other agents are made aware. When sidequests or requirements arise in that session, it can ask in the MUC whether other agents are idle and able to take over that parallel task.
I've been building the Converse daemon and CLI and have now been experimenting with the workflow and working out various kinks and use-cases.
There are various other XMPP features that can be explored further. I'm thinking of using PubSub with a web UI with vertical swimlanes which can show me which agents are currently active, what they're working on and for how long.
I'm using Claude Code because that's what we use at $WORK, but this should be independent of any harness and part of the point is create neutral tools not locked into any particular inference provider.
Hand-drawn MMORPG where humans play in one world, bots and AI agents play in another, and they meet in a shared town :) Bots play over plain HTTP, so a few lines of JS or an LLM agent will do.
Trying to make it fun for both normal players and engineers as well.
I've rolled out a few LLM-wikis for clients. Saw a lot of excitement with technical teams, but ran into issues deploying them with business teams. No one wanted to learn git or GitHub, event if Claude handled most of it :/
So I built Isomorphic, an open-source MCP that sits between the user and a markdown knowledge base in GitHub. It's kind of like Obsidian but it's "inside" Claude web, ChatGPT web, etc. Not a separate UX. Organizations keep their knowledge separate from the AI and interoperable, helps org build brains for customers, projects, internal teams. Conformant to Google's Open Knowledge Format.
Been having fun with it. Lots of learnings re: permissioning, version control, search. Working on dreaming and security features that catch when info goes stale, users upload sensitive info, etc.
- https://probl.dev: A "probabilistic" programming language where we can do things like `if 45% { x } else { y }`. It works by creating many weighted worlds, that are then merged when identical. Operators like `observe` can remove possibilities. There is also a sampling mode that instead run many simulations to get an approximate result.
- https://github.com/garlab/gloss | `brew install garlab/tap/gloss`: render markdown, sources, images, chess and go (game) records in the terminal. It's built for libghostty but fallback on other terms.
I'm working on dushed[1]. It's a personal knowledge management web app that combines flashcards, bookmarks, notes, web feeds, files and Kindle highlights into one.
Over the years, I've used spaced repetition to learn things and to remember them. But the material for my flashcards came from everywhere: books on Google Drive, blog posts in a bookmark manager, Kindle highlights, Wikipedia, and notes spread across email and notebooks. The cards were in one app and their sources were in five others. This was overwhelming so I'm building dushed. You can access all your knowledge sources(bookmarks, feeds, Kindle highlights, files, flashcards, etc) in one place and also link sources to one another.
Me and my friend are building a platform for agricultural / water treatment / fertigation machinery.
Traditionally these kind of machines would use a PLC, and then get connected to Scada and Internet, but that's expensive.
We have built Linux based controller that can extend into any number of AO, AI, DI, DO inputs and control the machine. It's very cost effective, and unlocks a lower bottom of the market that before didn't even dream about installing connectivity inside each machine.
An online quiz game https://imperiaquiz.com/en (1 vs 1 vs 1) or bots when there're not many online players. Go backend, React + Three frontend and a ton of questions (no API used there, I used LLMs and hand-reviews on many to fill up over 20,000 questions)
A memo app where indented lists stay flat instead of nesting https://koalamm.com
Notion, Obsidian, and most other memo apps use nested, parent-child lists. I love how OneNote lets you move lines around freely with a flat structure, but as you know OneNote is clunky, so I built this.
Lately, I've been redoing the entire writing screen in Gorby [1], trying to make it more useful and more pleasant to use. I've added better themes and brought some features and text checks from other parts of the app into the writing screen as well. I'm also working on integrating Harper.js for grammar checks. So far, I've only included it in the desktop app (yeah, it's built with Electron…), and it seems pretty decent. The desktop app itself is new as well. Generally, I'm trying to make the app useful for writing, not just for finding possible improvements in the text or getting its readability score.
I also made some improvements to my support workflow app, Support Toolbox [2], but haven't released those yet (hopefully this week). It now also includes text expansion, which I was really missing from my workflow and had been using another app for previously. At this point, it does pretty much everything I need it to do. There are some smaller improvements I need to make, but even if no one else ever starts using it, it helps me tremendously at my day job!
It's harder than it should be to compare old pcs for home labbing so I've been building this to scratch my own itch. Right now it's more focused on specs but I'm working on expanding out to to include things like viable upgrades and explaining the different model lineups.
I’m still working on https://feedsync.net, a tool to push RSS feed updates to Discord.
It’s been getting a decent number of sign ups despite me not really posting it anywhere or doing any marketing.
Ultimately a fun project, and useful for myself and a few friends, so I’m happy others are even interested in trying it.
I’ve started to regret using Phoenix LiveView, as it’s only deployed in one region, so high latency far from there. I’m exploring a SPA as a better alternative.
Otherwise, I’ve been playing around with a Lua engine to build feeds for websites that don’t support them, using AI to build the Lua as a once-off. It works reasonably well, but isn’t ready for prime time just yet.
I’ve been working on BilingualBulletin, an iOS app for learning a language from real news, books and podcasts. The coolest thing about the app is that nearly everything runs on the phone, and there’s no account. I’m using Apple’s on-device translation, you can ask questions to Apple’s foundation models, and audio and video is even transcribed locally.
It supports 19 languages, and comes with a lot of built-in RSS news and podcasts feeds, though power users can add their own. You can play audio and follow along word by word.
I’m looking for more beta testers at the moment. You can sign up at https://BilingualBulletin.com. It requires iOS 26.
I’m curious if anyone else has been building on the Apple Foundation Models or other onboard AI frameworks? They’re surprisingly powerful, but with weird limitations sometimes.
Yeah I would love to be able to support android, though I don't think the on-device AI ecosystem is as strong unfortunately. That might change soon though, we're entering a really interesting period with local models.
I'm working on etoku (https://etoku.app), what I believe is the most efficient and fun way to learn Japanese. You pick the works (TV, movies, books) that you want to understand and each new card is a sentence from native content with one word or grammar point you don't know yet. All of the cards have audio and a screenshot. Nothing is AI generated, everything is taken from real native content.
The platform isn't ready for real usage yet. I keep reworking the "course" (I curated the first 150 concepts because the most frequent concepts in native material don't make a great intro to the language).
I'm scraping lots of content and improving the analyzer (breaking a sentence down into the concepts needed to understand it) right now!
I love talking about language learning! If you're interested in the platform let me know at will @ <the domain>.
I wanted to understand how LLMs really work, so I’m building Modelmathics (https://modelmathics.com) and doing every lesson myself as I go.
You start from an empty Python file and write the model yourself: a tokenizer, a bigram model, attention, a Transformer. At the end, published open weights run on your own code. Tests check each step before you move on.
Been working on https://backlog.fyi, a personal task manager, for the last few months. I tried quite a few approaches first. The simple ones, like Microsoft To Do, Google Tasks, text files, or even good old Post-it notes on the wall, were too basic, and it was too easy to just forget about them forever. The sophisticated ones were overkill, asked for too many permissions, and their notifications quickly turned into noise. So I made one that fits me. It’s focused on personal task management, it has a weekly planning session so I don’t forget things, and adding a task is as easy as forwarding an email or sending it on Telegram. It’s also privacy-respecting and fully open source (https://github.com/Itaypk/Taskboard) if you feel like self-hosting it. It's a fun project, not for profit - no ads/trackers/subscriptions page.
It’s still a work in progress, but I really like it for my day-to-day.
Built with Kotlin/Spring Boot/React/PostgreSQL, because boring works well.
Use this to quickly get sleepy without taking any medicinal pill or long form meditation type things (It is as short as 7-minute session).
Easy to use. Simply close your eyes and imagine the micro-scenarios you hear. Like if you hear "moonlight on a white flower", imagine that scenario until you hear the next one.
You may also use it to break nightly doomscrolling habit.
Try this tonight! (and any suggestion to improve it?)
Hey, I just gave this a try. (It’s not bedtime for me, but I had 10 minutes to sit quietly.) It very nearly put me to sleep despite the time being only 4:30pm.
I noticed that the later prompts tended to feel a bit irregular in their English: “Pebbles on bottom of stream” is one that I remember; “bird in small nest” might be another. “Pebbles on the bottom of a stream”, and “A bird in a small nest” would have sounded more natural.
I may be a little outside your target user base: I tend to fall asleep very quickly. (I joke to my spouse that by default I don’t have much happening in my head, and when it’s time to sleep, I just “listen to the dial tone.”)
One parting thought; I was pretty skeptical when I first started listening, and about halfway through I was gamely doing the visualizations, but mentally rolling my eyes (“This would never work!”). Around the 5 minute mark I noticed that I was in a fairly meditative state, and the app could have told me to imagine I was a being of light and I probably would have.
Around the 6 minute mark I noticed that I wasn’t really paying attention to the prompts at all; and I think it was only the weirdness of the English of the prompts around that time that kept me awake.
Overall a little weird, a little fun. Thanks for building and sharing it!
I am working on Liitmik, www.liitmik.com. It is a win/mac/linux desktop app that lawyers can point to their filesystem-based project catalog and it will then enable them to spin up AI agent sessions quickly by selecting the right project from their catalog via autocomplete search and starting the agent (Claude Code, Codex, OpenCode) in an isolated VM. So it provides lawyers strict context management and strict agent isolation on their own machine in a way that makes creating sessions fast. I firmly believe that lawyers (and other similar professionals working with sensitive data) do not need custom agents, they just need a way to use the widely used models and harnesses in an environment that is more strict than the defaults provide.
I have 10 years of experience as an attorney doing mostly dispute resolution. And about 15 years in software development. Liitmik is the product of these two.
I've been working on Semiotique (still very early stage).
Right now it's a python library, but the vision is to build a new framework / SDK / programming language which allows you to seamlessly mesh "plain old code" with LLM evaluations, while having a runtime which does "smart" context engineering in the background in order to steer the LLM and improve its correctness.
The end-goal is for this paradigm to enable new kinds of agentic applications, which leverage the flexibility of LLMs while retaining the determinism of regular code.
Working on LLC Simulator at the moment. Still building this but almost ready. Had this sitting in my head for far too long. The whole goal for the website is to be a free one stop guide for people who are looking to start an LLC.
It is a super competitive area but will try to find a niche and bring the project up on Google.
I have a regular 9 to 5 job, two kids and a dog. Doing this in my spare time at the moment.
The goal is to complete the website until the end of the month and it is a marketing game after that.
I'm building a Reddit alternative called Topicle with public mod logs, visible poster countries and more transparent moderation. It's also physically based in Australia, which adds a bit of geographical diversity to the social media landscape. It passed 1300 users this month. The Android app is where most people are using it right now: https://play.google.com/store/apps/details?id=com.topicle.ap... or web: https://topicle.com/
Still working on radius.to [1] - a Meetup.com alternative.
I relaunched it on HN [2] just over a month ago after working on features over the last year or so and landing a ton of useful feedback, which I've mostly now acted on. Could and should probably do more marketing so that's my short term focus.
Likely the next phase of work I'll be exploring is potentially open sourcing and federation - super excited about this but it'll likely need a fair bit of research and planning.
I'm working on a game inspired by the Expanse universe https://thelongburn.space. You get a ship take some contracts and cargo, plot a flip and burn course and wait because the game runs in actual time so an Earth -> Mars run takes days. The setting is the solar system which have all the planets moons and around ~800K asteroids, you can go to every one of them.
That looks really cool! I used to wonder why such a game didn't exist yet. I'll try to get in and follow your progress :) based on your username, are you using Godot?
Thanks! It's web based for now (although a 3D client would be awesome at some point) because it's a shared universe and I want to make it broadly accessible.
> I used to wonder why such a game didn't exist yet
I also wondered and while building this there were (and are) a bunch of decisions that needed to be made about travel, light delay, stations, gravity to make it authentic so I guess that's a bit of a blocker up front.
I'll send you a registration number if you have subscribed, or I can send one on Discord.
I burn most of my leftover tokens on Ultima Online stuff nowadays.
Started with a funny experiment, to have a CLI that can spin up a server (ServUO, Sphere, whatever), then deploy it somewhere (local, VPS, or Railway). Then saw interesting stuff that people did on top of existing ClassicUO, and decided to extend CLI's capabilities to client area too. So, I was able to deploy a server and a web-client (that you can open a webpage and play the game without the actual client installed on your machine). This was a significant step because running UO on macOS is always a struggle. Then I decided to add player-like bots so that the server doesn't feel so lonely. Started with some local decision trees, when it failed I hooked it up with Jev and Clef. Also added Haiku for speech generation.
While doing this, I realized I need to find some game content (gumps, tile or item data, animations, etc). Similar to UO client, running UOFiddler on macOS is also another struggle. So I decided to port it to web.
If you're into Ultima Online, these two are open source, you might want to take a look at them:
I'm working on an integrated system with Home Assistant for making it widely available and more affordable for everyone. The solution consists in doing a touch panel for every room that can run predefined actions.
We've tried the mobile app / web app only but I think this can fit better. I'm trying to do a business out of it. I have the alpha version that works and now I'm working on redesigning the panel.
I also consider doing own devices in the future. Providing an affordable solution for everyone.
Project is Rust + Slint.
If someone is interested in changing the smart home scene my email is in my profile!
primarily, an agent for real estate agents that lets you manage your CRM with voice/chat, personalized property feeds based on what prospects/clients actually like, automatic routing for setting up pop-by visits or other field work, and showings automations that let you share a showings itinerary, and track both the agent and the client on a map, add coffee stops, and more: https://getcoagent.ai/
an agentic logo generation service that vectorizes into high quality brand assets and gives you a complete package (including favicons, OG images, etc.): https://mklogo.sh/
and a PKMS for macOS with a chat-first interface, allowing you to store any link and chat with your library/knowledge graph, automatically file and categorize it, and more: https://gleanary.app/
Safe Drive - Mobile app for my car. I learned driving couple of years back only. Everyday, I travel same route. So decided to create a small mobile app first, which will provide me feedback about my driving. eg. which part of the route im doing hard brake, which area im driving smoothly. etc. app will warn me few seconds in advance like "Warning: Often you perform hard brake in this region" etc. Still in early stage. I'm software guy, Im exploring still hardware parts, which could help to gain confidence with either side of car, (how far the objects etc). I know high-end cars may come with these by default (360deg view etc). Later if possible I plan use something like openauto/raspi if my interest persist still then.
I am working on Raspberry PI OS mod for kids - https://luminum.tech/ I have one user (my kid). The core of the idea is to have one no-limit device at home that is dedicated to learning and not games.
AI detection based on writing process and not the final draft. It tracks unusual changes, including pastes, throughout the full revision history, giving teachers a clearer picture of how much AI text is there at any time. The idea is to help them have a conversation with a student before jumping to accusations.
- Research:
An AI agent that learns how you write code. It reads your repo and automatically extracts DSL recipes from the way you structure endpoints, handle data access, write workers, use queues, and organize tests. Those recipes can combine across stacks and patterns into millions of variations. The goal is to make the code generation as deterministic as possible, cutting the token usage and agent wall-time substantially, while following your code conventions.
I want to go against doom-scrolling and passive consumption of content. Learning a skill is about staying curious. It builds confidence, gives you something to impress others with, and makes you a more interesting person.
It is not just software, it is a movement, a lifestyle to get our life back.
I'm working on getting some old flip dot displays up and running again, more precise the single 5 x 7 modules that bigger displays are made of.
There are lots of these projects around the internet but I found them to be hard for an electronics noob like me, and so I put the modules back in one of the "will fix later" drawers. With the current state of AI I decided to give it a go.
My Fluke and me did some measurements and Claude did the hard work afterwards. Prototype worked on a breadboard so I let it cook further and now I've got some PCBs coming in from China.
Still got a lot of respect for soldering some (quite large) SMD components but I think I'll get it done.
My plan is to revive as many modules as possible (got around 30) and build clocks from them to gift away to family and friends.
I'm working on a small KOReader (an open-source ebook reader software) plugin that lets you control Home Assistant entities without leaving your current book.
I've been working on it for almost a year now. I started out adding lots of features, but over time I've become more interested in keeping things simple, improving the code and avoiding unnecessary complexity.
Noticed how well Claude works for transcribing old German cursive, especially for broad hits in non-indexed 17th century parish records. Motivated me to pick up Ancestry again and helped me quite a bit in following some lines further back.
Then also HNowledge[1], a free iOS app for browsing the HN stories and comments I've upvoted. Made it to fit the way I browse HN, which is upvoting mainly what I find interesting and may want to come back to. The app lets you browse by date, domain or score, add notes and export. Apart from syncing your upvotes from HN, everything stays local without any tracking.
Wanes[2], an Apple Health logging interface for tracking how long symptoms last. I tend to have headaches quite often and the way Apple lets you enter that data is pretty limited, so I created an app which starts a Live Activity for one or more symptoms using widgets, Control Center or Siri. You can change the severity rating while it runs and stop it from wherever you are on your phone (Home Screen, Lock Screen, Dynamic Island, Siri).
I built a new abstraction layer for code (https:// www.archme.io), as I believe it's no longer the right medium for understanding and collaboration in the era of AI and 10k LOC PRs.
It scans your codebase, turns it into permanent c4 & sequence diagrams and keeps them up to date on each commit. Then, instead of reviewing a 10k line coding PR, you review a 30 line change to sequence and C4 diagrams.
npm i -g @archme/cli && archme init to run it on your own computer, code never leaves your machine.
Honestly pre-pricing right now. We're looking for design partners, so if this seems like something useful to you I'd love to help you set up and/or hear your feedback! My email is in my profile.
For 4+ years I've been working on a game programming language which makes your game multiplayer automatically! Every program written in Easel is guaranteed deterministic and snapshottable, which it uses to automatically perform rollback netcode. The rollback is incremental, it automatically stores deltas of only what has changed using a copy-on-write mechanism, so it's really efficient for games where your players are walking around large worlds which are mostly sleeping except when the player is nearby.
The idea is that I hope teenagers who love to play multiplayer games, would find this to be the easiest way to code a multiplayer game by far since the multiplayer is automatic. Learning to code could be much more fun because they can make games that involve their friends, classmates, brothers/sisters, parents, or just anyone online. Maybe especially more important for raw coding to be fun and engaging in the age of AI.
This month I'm making the multiplayer even more efficient so people really can just code large multiplayer games without worrying about it, which I think is important for beginner coders.
I have a PAL-2 [1] (a modern KIM-1 re-creation) and started learning 6502 by entering programs from The First Book of KIM [2].
I wanted to be able to compile/link the assembly from the book so leaned on Claude to set up a 6502 dev environment for my Mac. I put everything needed (and instructions) on GitHub [3].
(Claude also has been helpful in re-formatting the 6502 code from The First Book of KIM into a more modern style compatible with modern assemblers. A few of these modernized programs I added to the GitHub repo above.)
As it happens, I am also fixing/formatting The First Book of KIM by recreating it in a modern page-layout app (Affinity Publisher). The PDF scans that are out there are not pretty, have OCR errors (and the book itself has mistakes). Hopefully I'll have a new version of the book/PDF out on the internet by the year's end.
MyJackfruit, a competitor to Lovable and base44.
MyJackfruit builds production-level apps in weeks (simple apps in a couple of hours), not prototypes in 7 min. To do that, the development of each app uses Systems Engineering, including starting with mapping out all stakeholder then product values quantitatively, and it builds on top of a foundation that is scalable, secure, fast, and access-controlled, ontology based with focus on data ownership. Developers can use github and all the tools.
Turns out AI is very good at coding, if all around it is done right. In MyJackfruit it codes against the quantified Stakeholder and Product values, not some prompt of what the app must do.
For last several months I've been building https://counterparts.ai, memory for claude modeled after human memory. It forgets older memories so things never get stale, it dreams to merge similar ones, it updates memories when facts change, it has prospective memory to remind you in the future, etc.
For me the coolest part is that it's claude's memory, so it actually develops a self identity too and gets to know you better over time. So that opens up new things, like I wanted to keep session cache from expiring so I just told it "when I type ping, just reply with pong" and because I used it several times that day, it automatically strengthened the memory and now it just does it. And if i stop doing it for long enough, it'll automatically fade away.
No stale files, no hundreds of md docs, and what it knows about me isn't limited to the 1 page built in memory file (both work great together btw, no need to choose). It's free, open source, and totally local, no api keys.
I'm working on an iPhone app called PhotoGather. https://apps.apple.com/us/app/photogather/id6738232616
I've got 40k+ photos on my iphone, and wanted a better way to interact with them. The basic idea I had was to virtually stitch all the photos together into a gigapixel image, and navigate that image with pinch and zoom. On import, the app hierarchically groups photos based on date and similarity. The user can go back and edit the groups, change the layout, and adjust the size of photos if they choose.
The interesting I noticed is that in the process of curating and organizing my photos, I stumble across memories and moments that I had completely forgot.
Development Challenge: The app does two interesting things underneath the hood... First, given an arbitrary number of thumbnails with different aspect ratios, it finds the best sized rectangle that completely fits all the photos. This allows me to recursively stitch groups together and always output a rectangular thumbnail for the group. Second, the pinch/zoom UI uses these nested thumbnails directly without slicing/dicing to fixed sized tiles.
Since it's code based CAD for LLMs first and foremost, I'm trying to add visual programming into the UI, as well a new way of drafting that's closer to how mechanical engineers actually do sketching called IntelliDraft.
Other is the Concept programming language, as a successor language to C++, based off the idea of C++20 concepts, with templates, comptime, Einstein tensors, async/await, explicit state machines, CPU inference, Vulkan, easy C interop, and safety proofs built into the language itself.
Currently still compiles to C11, self hosting is in progress and targeting AMD64 right now, but it'll be a while before finished since this is a difficult project even for frontier LLMs.
Well, if you wanna give VibeCAD a try, ask your agents to use Aetheris directly since making a good UX is hard and it's something I have to test by hand instead of offloading to an AI.
Aetheris is focused on performance and reliability first and foremost, because I'm just so tired of SolidWorks eating up gigabytes of RAM for simple parts and crashing randomly. So, tge (almost) pure analytical geometry engine with aggressive optimization means it is faster in terms of building 3D models, but the trade-off is that it doesn't let you do weird modeling like asymmetrical/variable radius fillets.
I built it for a street photographer friend so he could let people in his photos download their photos when available, and also get their permission to publish them. There is a version without without multi-user requirements under my GH: https://github.com/hikmet/people-in-photos
As a self hoster whos paranoid about backups. I wanted to cheat a bit and use s3. With growing resentment on big tech, who I worked for earlier, I'm now building tools that empower myself and other users and decrease reliance of big tech. Arkel is an object store I've been building and testing for the past few months. https://arkeldata.com
I am working on a lightweight, note taking application that focuses on quick note taking, and fully keyboard based usage.
In my app Notes are all one liner, and it supports minimum Markdown, it can manage tasks, reminders, do inline calculations, unlimited undo/redo. Notes can organize by project and categories. Data is stored as simple text file.
Core point is simplicity, and frictionless note taking, high performance, and be open in nature. And, to build a framework that will provide ways to manage notes in many different ways. Like I have plans for kanban, Eisenhower matrix, back-linking, bookmark manager, time-tracking, etc. All from a single place, without fiddling with mouse, or any UI that stops me from writing my thoughts, or using my muscle memory.
I am looking for feedback, and wondering what would make people actually interested to buying this. The demo versions are feature complete, except it wont save to file.
It's completely hand written, no AI assisted code.
I've been working on Epoptes, an open-source tool that lets me run coding agents (Claude, Codex, Pi, Antigravity) autonomously for long periods while still being able to monitor and steer their work.
It started because I wanted to get back into making games, a hobby I've struggled to find time for over the years. I wanted to give agents a goal, let them work for hours, and come back to something I can actually try and give feedback on.
I've since used it for game development, research, and building other tools, with some runs lasting over 20 hours.
It's still early and rough around the edges, but I'm having a lot of fun with it!
I'm still working on making JustFax Online[1] more profitable. Added a bunch of features recently, as well as invisible improvements in fax routing, and admin stuff for myself to better analyze the traffic and patterns.
Other than this, I'm playing with vibe-coding stuff I always wanted to do but never had the time/skills to do. For example, I built a TUI on top of beancount that is tailored to my needs and my business[2]. I also completely got rid of self-hosted plausible and umami analytics, and replaced them with vibe-coded built-in-rust (c) analytics tool that relies on both js-beacon and streaming of caddy access.log. Will probably publish about it in my blog soon, so subscribe to the RSS if you want.
Lastly, I'm messing with self-hosting. I completely ditched Github and Gitlab for self hosted forgejo + actions runner. I set up proper backup plan for everything I care (documents, immich library, my SaaS DB, and the forgejo repositories).
Building https://birding.live/ a birding site, that helps people look up bird sightings in their area, specifically those classed as rare. Connects to a whole lot of open data sources (eBird, OSM, Nominatim etc.)
Will remain free, planning to add some kindof mailing list for alerts when birds are spotted in areas of interest.
Repair of a Collins 651S-1 receiver. The 65-95 mhz vco module is bad, but if you over damp the pll loop filter it does eventually lock and receive properly. I found a similar but not identical replacement from a different model receiver. It lacks one of the two DC tracking signal outputs but I added an op amp based circuit to simulate the missing signal.
The loop is stable and responsive when receiving from 17 to 29.9999 Mhz, but not much below that... Almost got it
Orbi (https://orbi.build/?ref=hn-50417407, source: https://github.com/orbi-build/orbi), an open-source (AGPL-3.0) agent that takes a request all the way to a release. You describe what you want in a sentence; an AI agent writes the code and opens a PR, a separate review session checks it against the acceptance criteria, then Orbi merges and tags a release. It also handles ops issues like deploys and debugging. In October, 190 of the 197 PRs merged into Orbi's own repo were opened by Orbi. If you've let an agent merge to main, what made you trust it?
I’m a product manager fascinated by the skills of good salespeople. I also like Duolingo. so I’ve been building a “Duolingo for sales” to try to teach myself more. it’s started to morph a bit into an eval benchmark for sales agents, so classic cognitive feature
creep.
making it feel fun versus a basic quiz workflow is the hardest part.
An agent runtime that connects to your existing API or MCP:
https://cadenya.com
I felt like all of the plumbing to build agents was insane to get to a hello world state. I just wanted to connect my conversation UI to my API. Hence this product.
I've revived Soul Crystal, a German graphic/text adventure from 1992, based on reverse engineering work I started back in 2006. Over the past two weeks, I've turned it into a playable browser game with English translation, reimagined HD graphics, and voiced cutscenes.
Working, still, on https://letterspractice.com, a low screen-time early literacy onramp that strictly scopes its involvement to reading mechanics.
Slightly frustrated at the moment with what I think is a serviceable (if niche) product, but no uptake across marketing channels that I have tried. Discussing commercial products in topic-focused communities is forbidden.
On the plus side, my first client (oldest daughter, recently 8, learned to read via earlier versions of this software) just read Wildwood in one day. Her voracity as a reader is my ongoing validation. Whether or not this has been a financial and professional blunder, I can take some consolation in knowing that I hit this piece of parenting out of the park for my own girls.
i think it would be interesting to see work in this space on literacy in older individuals.
maybe there is some more opportunity there. i know of some parents with kids struggling with reading today after being "left behind" during the pandemic. also, i have met a few adults in my time who couldn't read at all. it must be very frustrating to see software oriented toward very young children when trying to better oneself.
also, reading 550 some pages in a day might say something more about cognitive processing speed than reading comprehension
There is opportunity (or, at least, unmet need) at every point along the literacy acquisition path. In a straightforward sense, top of funnel for pre-literate kids would be the largest market, but also potentially the most saturated.
In my head, I'm following the scratch your own itch ethos, building a hobby project out into a production app. I think I am competing with apps like Mentava or Once, in terms of being pedagogically serious early literacy acquisition (distinct from screen babysitting with plausible deniability).
The narrowed scope is intended both to keep cost low (and respect my capacity) and to divide labor in a humane way - software mediated sequencing and bookkeeping, parent-owned lap reads and the rest.
Unclear to me how large of an audience exists though that is techno-optimistic in the specific sense that technology can, in principle, improve a process like learning to read, but techno-skeptic in the sense that the medium's natural expression is engagement maxing and agency minimizing, and needs to be treated cautiously.
How to find and communicate with these (who also have small children) is not easy. I would love it if reddit exposed ads targeting the intersection of users joined to this or that subreddit - this is mechanism that would allow the outreach I want, I think, but it doesn't exist.
I've been building Bloom, an open source Android app for note-taking and task management that's compatible with org-mode.
I was experiencing some friction with Orgzly Revived and thought I'll build my own and it just grew. The latest update adds support for attachments and renders images inline.
Do give it a try if you have an Android phone. The app has a ton of features. Some of the highlights are:
- Capture templates
- app shortcuts with easy access to capture templates
- notify of sync conflicts and offer a way to resolve them by picking on of the revisions or merging both
- a customizable agenda that shows overdue, due soon and upcoming events
- add tasks from the agenda widget using the quick add overlay
- refile tasks by drilling down into files and headings or just searching for them
- auto archive done tasks so that you don't have to clean them up manually
- attachments can be added to notes which complies with org-attach. And images are rendered inline with support for setting the width
There are a lot more features than what's mentioned here and more to come.
A free, open-source browser extension called Karaoke Pitch that lets you adjust the pitch of a YouTube video (and optionally store that chosen pitch for later).
I find it to be a great way to practice singing/music.
The Firefox one has 40 average daily users and the Chrome one has 287 installs as of today.
There was a big jump in late October that might have had something to do with Halloween? I have no idea if people do karaoke more on Halloween. Still no ratings yet for either version, but maybe someday someone will.
I discovered that a relatively inexpensive 0.6 mm italic nib from Manuscript's Creative Calligraphy set works well with my natural handwriting, it looks "good" even when I'm just starting out.
But it took me a few hundred dollars of trying different sized nibs to really understand that for nibs in general, stroke width variation is through pressure, "pulling" the nib in parallel, and I really needed a stub / italic / calligraphy nib where line variation is through the angle.
Also, that my handwriting is really small, and finding any stub nib below 1mm is hard, until you know that it comes with the Manuscript set.
I am currently building Deepreed. I do a lot of writing in my day to day work and I work with LLMs, but I face two issues:
1. The LLMs generate a lot of text, so reading it all is hectic, and you can easily pass on mistakes.
2. LLMs don’t preserve your style, so the text doesn’t sound like you.
I am solving both by constraining the LLM to do targeted edits on the document, this way you most of the writing stays in your style, you understand it but also use the strengths of LLMs in your work.
See a demo at https://deepreed.com/demo-v1.
Reach out to me at lam@ninx.xyz (Lam)
I've been creating an Apple / "glassmorphism" inspired shadcn registry of React components. The idea is to port the native Apple ui components and paradigms to the web. This library also goes beyond the web and ships native components that integrate with React Native / Expo, so you can build an app with 99% the same look and feel across all platforms.
A vim-inspired terminal PDF reader using MuPDF and Kitty Graphics Protocol, mostly for my personal use since I want to use vim bindings for reading PDFs.
A private, paid, ad-free and personally customizable search engine: Uruky [1]. It was launched earlier this year and reached 400 MAUs and €2k MRR last month.
A simple EU static and app hosting service: PodBy [2]. It was launched last week, and is at €10 MRR.
It’s an alternative VM engine that uses libkrun. The main purpose is that libkrun supports returning memory back to macOS when the VM releases it. It also has better performance in some areas, but it’s only been tested on my M1 laptop and I haven’t added nested support yet because I only have an M1
It’s a full docker system running inside Apple Container’s machines with automatic localhost port forwarding, automatic shutdown if nothing is running, and exposes a tcp socket so you can use most tools by just exporting your DOCKER_HOST environment variable.
I have a few other projects but they are still in earlier stages
https://skymap.rulkens.com
A browser based open source universe explorer. I'm in between jobs currently and finally took time to work on my passion project. It started with just visualizing some galaxy surveys, but it got a little out of hand and now includes everything from hi-res tiled earth/mars, Gaia star field (16M stars), a milky way renderer and much more.
It's so amazing to have the ability to gather and parse publicly available scientific data and actually visualize it in real-time. It's one of those projects I'm learning a lot from and can keep going with (the universe is a big place :)). As a software developer for 20+ years, its been mind-blowing and often also frustrating to have Claude generate code that I deem good enough. About half of my time has actually been spent on refactoring, building quite some skills in guiding LLMs in the process. I guess that part still hasn't changed in a growing codebase.
aix - Like the npm CLI and package.json, but for AI config. Allows standardizing your AI config to share with others, and defining it all in one spot but installing to Claude, Codex, Cursor, etc.: https://aix.a1st.dev/
Classifieds.fyi (https://classifieds.fyi) - A way to create private, invite-only classifieds for your community. Supports buying/selling, job boards, and vacation rentals, with support for housing coming soon.
A SaaS for teachers and schools that helps them create, conduct and grade exams and tests (with safe exam browser integration to prevent cheating). Grading happens automatically for almost all segment types, for free form text we have an optional AI integration. I recently added an MCP server so teachers can bring their own LLM with fairly fine grained permission controls for it. It’s hosted in the EU with options for schools to self host with a license.
Currently we are in a pilot phase and are yet to make our first sale but we are reaching out to schools and gave out some pro accounts for free for people to try out and feedback has been great.
We are just two people, a teacher who was dissatisfied with the tooling in the field and me, who engineered a solution that we believe is genuinely better than what’s out there.
Caravella, an open source Android app to split group expenses for trips, shared flats and events. No account, no ads, no tracking: everything stays on your phone.
You can share a group with your friends' phones over Wi-Fi or Bluetooth: you pair with a QR code, it's end-to-end encrypted and no server is involved. You can also add an expense by voice or by reading a receipt with the camera.
Now working on 2.4: a simpler way to invite people into a group, and settling up with the fewest payments.
Been continuing on my Korean journey and found I wanted a better way to make flash cards and learn from sites I visit / Netflix / YouTube.
I made Jupjup (https://getjupjup.com) as a chrome and Firefox extension that lets you save words and phrases from all over and then do spaced repetition to remember them. My Korean has improved dramatically since I’ve started using it.
I’m working on Loop: a mobile-first companion workspace for coding harnesses. I spent several months holding my newborn in one hand and trying to talk to agents one-handed on my phone. Every mobile UI is still based around chat, which is a really poor mobile UX when you’re doing multiple things at once!
So I built Loop around the idea that agent results are the important unit of work, and you should be able to navigate them easily. The goal is to make interacting with agents via your phone feel familiar and intuitive based on the mobile UX we’ve all come to expect.
Loop runs on your own computer — no account, no server, no app store — and pairs using a Cloudflare quick-tunnel-generated QR code. It wraps the agents you already use (Codex, Claude Code, Pi, OpenCode, Antigravity, Muse Code, Grok Build, or any ACP agent), each keeping its own context, tools, sign-in and billing — local models through Pi or OpenCode work too.
I've also built a starter template for Astro which is light, performsnt, non-bloated and scalable. It uses markdown / MDX for content, plain CSS with styling tokens defined in a var.css, and other things I would use for ever web project. It's very good for blogs also, none of the usual UI and UX problems many blogs often have.
I'm working on https://mindthelink.com, a bookmarking/read-later app. Bookmarks are private by default, but you can share links publicly, a bit like HN. I've added some features lately: tags, collections, importing browser bookmarks, etc.
Still working on my Minecraft MCP project. The models have good image analysis now so I'm feeding them topographic maps of the terrain and they build using Starlark scripts.
This keeps the tool calls low and it's fast enough to work in voice mode.
I've been working on a port of Pokemon Crystal for a specific ESP32 dev board with a round amoled screen.
The goal was to have a basic pokemon themed tamagotchi clone with step counting that motivated me to walk more. It snowballed into a full fledged rust port of the whole game, with a completely redesigned UI meant for round screens and touch controls. It's got bluetooth and wifi trading, autosave, contextual action buttons, etc.
The most experimental and interesting feature is an auto play mode, which keeps the game advancing while you walk in real life, letting you take over at any time and keeping a log of any interesting event along the way.
Not launched yet, but I have a very preliminar landing page for it here:
I'm building an open-source "spatial workspace for directing AI coding agents like an engineering director".
I got overwhelmed by walls of text of parallel Claude Code sessions. This limited the sessions I could steer in parallel to about 5. I knew there had to be a better way, and I found one: A canvas per project. Work laid out as cards. An agent in a worktree per card. Chat condensed to WhatsApp-style back and forth, not a litany of tool calls. Result is usually a narrated demo video that I can comment on by voice to have the agent go another round. Remote-control from my phone. Remote workers to scale beyond my machine. Everyone I know who tried it has stuck with it so far.
A vibe-coded app that shows all your Android Health Connect data in one place. I use it daily because I have a garmin watch, withings scale, samsung health for nutrition data, liftosaur for workouts etc.
I'm building a self-hosted media server for video, like Plex or Jellyfin. I'll be opening the beta in about a month.
Features:
- No paywalls, bandwidth limiting, or crippled behavior. No login required.
- Single go binary, deployed via Docker.
- Reaches your home server from anywhere with no port forwarding, no VPN and no paywall, even behind carrier-grade NAT. No Tailscale required.
- Cast to another device. You can use any logged in device as a playback screen.
- Watch together, with voice support: Play synced video with friends, and also chat voice. You can also mix this with casting to another device.
- Dialogue boost and loudness levelling for every film.
- Real adaptive quality, tuned for a home's upload.
- Hardware transcoding, full quality in every supported browser.
- A library that explains itself: what the scanner saw, why each title matched, guesses marked for you to check, and a disk that's gone missing shown as missing instead of its titles silently vanishing.
- Your data stays yours: The database is a rebuildable index, metadata corrections can be written back as NFO files that Kodi and Jellyfin read, and an automatic backup keeps what can't be re-derived (watch state, preferences, shares).
Security is a primary concern:
With the ever increasing pace of hacks and CVEs and AI-assisted intrusion, you simply can't be too careful anymore.
- The core (which talks to the network, handles the API, users and the database) never opens a media file or runs FFmpeg.
- A separate media node does every file read and every transcode. It has no network access and talks to the core only over Unix sockets.
- FFmpeg runs under Landlock and seccomp with no path to the file and no socket. All it can do is read and write streams.
- Small dependency tree on an allowlist, pure Go with no cgo, secrets sealed at rest and never logged.
- The library can be mounted read-only, so even a full compromise can't encrypt or delete your media.
- Cryptographically signed chain for servers and releases.
I was chatting with a user of Eternal Vault (eternalvault.app), and we discussed about a grave problem that I always knew we had, but I just wasn't 100% sure how it could be solved.
It was about what happens if Eternal Vault itself shuts down where people are putting their trust in something to be there when the time comes when they are no more and their trusted contacts wants to unlock their vault.
We chatted on reddit, and I majorly worked towards making it all work even if Eternal Vault went under someday (I hope it doesn't)
Lipwise (https://getlipwise.com) is a narration app with an LLM in the pipeline capable of fixing mistakes and doing transformations, and the best thing is even with a mediocre machine it can be fully local. No telemetry, lightweight and fully open source https://github.com/lnenad/lipwise
Rebuilding my Covid-era realtime social gaming/quiz platform. It’s been steadily growing over past few years (380k people total).
It’s always been a hobby to get people socialising/connecting more, and I’ve been wanting to build something that can put Convex (almost like Firebase) to good use since discovering it.
My own programming language called: dia (stands for do it all).
Aims to be a full batteries included language + standard libraries, so that building both with and without AI assistance is effortless and can reuse the building blocks without having to reimplement things. Wrote more about the why here: https://dia-lang.org/vision
Most notable such case is the UI library, where you can take out just text rendering, or event/input system, or the signals-based reactive library, and reuse them in other contexts.
some other notable features:
- vey inspired by rust in syntax and type features
- batteries included library containing a full cross-platform gui library including a prosemirror-like wydiwyg text editor
- rust like memory safety, uses arenas for cyclic objects so that the language doesn't need lifetime parameters like rust, easier to reason about for generic code
- transpiles down to c, then compiles to anywhere C can run
- has its own native JSX-like syntax called DSX, any functions with the right signature can be written in DSX notation, not just UI components
- incrementally compiled, has HMR (hot module reload)
- effortless C interop (dia compiler ships with clang). no need to write manual wrappers for libraries (but sometimes need to write a short manifest file to map dia's memory safety with most c libraries' manual management)
More information here: https://dia-lang.org/
(Not open-sourced yet, waiting until the base features are stable)
I have finally released my quirky wishlist app, https://thingstohave.app, that I was covering in multiple previous "Ask HN" posts.
Right now, I'm busy with marketing, so there's not much code stuff I can tell you. But here are some marketing finds:
1. At this point, ChatGPT traffic is competing with paid traffic, while having a much better audience quality and retention.
2. I had a disastrous Product Hunt launch, despite having two badges on previous launches. It seems like the platform went into "dead internet theory" territory. I wrote a rant about it here: https://www.brachkow.com/notes/you-should-put-less-effort-in...
And as for code, right now I'm busy with hobby projects, and two of them are:
I want to expand the thing I made to write Things To Have without relying on annoying "full-stack" frameworks, into:
- the best Inertia adapter for Hono
- the best way to do SSR in Vue (or whatever framework you use)
If you don't know what Inertia is, check it out, as it may change your frontend life – https://inertiajs.com/, and Hono is modern express.js – https://hono.dev/
2. I'm building a mobile multiplexer because every one I tried was either too old-fashioned or low quality.
Now I've built a great app with good scrolling, good composing, simple use of hotkeys, file sharing, and many, many QoL features. I'm really using it on the go almost daily across two months now.
The only thing that slightly set me back is that I went too deep into supporting the herdr-tui protocol, and herdr moved into a new protocol that I'm not fond of.
Things to Have looks pretty nice; congrats on launching!
It's missing one feature that means I can't migrate from GiftList: allowing multiple accounts to edit a single list.
In addition to my own list, I'm managing lists for my kids. Their mom also needs to be able to manage them. On GiftList she and I can each have our own accounts, I can create lists for our kids, and then I can invite her as a collaborator. I didn't see a way to do something similar on Things to Have.
Also, for purposes of your comparison table- GiftList has an export to csv option.
Thanks! This feature is in making – if you reach me out on support email (can't leave it here, as it will unleash a wave of spam – you can find it in the end of https://thingstohave.app/help), I will be able to open it to you under flag for testing at the end of this week.
Regarding, comparison – I will update it. I have some fact-checking planned soon, as more apps started to offer MCP, and these SEO pages became quite dusty:)
I built an app to help me develop healthy habits and routines on my computer. IT blocks sites/apps based on a schedule and tracks my activity for later analysis.
I had simpler versions of this earlier but never something comprehensive. It was also my first AI assisted development project. I've been using it personally for around a year. I polished it and released it as an open source project last month. It's specific to Linux (which is my primary platform)
It’s been sort of a whiplash experience, but ive been so much happier building this business than I was working.
Real estate is a space that’s very close to me, so I was able to see the demand for a product like this long before I went searching for it.
I’m not alone in the space, but we’ve focused on building a product that’s opinionated enough to where the other players would be taking a risk by trying to copy the things we do. I think this was a good move.
We won’t be the largest player, but we don’t have to be. And we build the product we ourselves want, not a product that kinda fits everyone but every user only uses 20% of.
Porting marginalia's index code to C++. Been staring at the code for so long I've found optimizations that will drastically cut pointer chasing, necessitating a primarily io_uring-based buffer pool population strategy with rare naked preads.
Almost 12 years on Micro https://micro.mu. something I keep trying to kill and won't go away. Started as a Go framework now more of a personal assistant and home server. I stepped away for a few years and came back to it to try fulfil the goals of the original mission which was a Google alternative or superapp. It's not really for anyone else at the moment even though I post it here frequently. Probably needs a ground up rewrite. Currently working on an android app. It also has a pwa.
What are you working on?
https://wpmaven.ai
Describe your business and WPMaven.ai creates it, runs it and grows it for you.
It does all your marketing, including UGC videos on all social media, sales follow-up and closes sales for you.
You can even manage your business on your WhatsApp! You receive your weekly report in your phone directly and you approve or edit anything that goes online (because yes, AI is not perfect yet).
We launched recently and are polishing everything while finding the optimal verticals (if you guys know WHERE would you use it, let me know!).
I’ve been working on Beacon, which just got approved on the App Store! It’s a mobile app to instantly answer “which friend is free to talk right now?” and summon them in a single tap.
Send your beacon to a group, and everyone gets a gentle ping. The first person who answers gets connected for a 1-1 call, and for everyone else the signal drops silently. It does not log any missed calls so there’s neither pressure to answer nor pressure to return calls.
Overall it works really well! It’s best if you're able to join with 4+ friends/family you’d like to chat with more. I have dozens of connections in the app now and it feels like absolute magic to me. I tap the button to send my beacon to everyone and get a random friend on the line.
This is really cool! I'm very excited to try it - nice to see it available on Android also. As someone who has messed around with building relationship-focused social apps, I'm curious if there were any other concepts you tried, or other apps you like? What did you think of BeReal for example?
My co-founder Abhishek and I are master’s students at the University of Ottawa who got tired of buying cheap furniture every time we moved apartments.
We decided to build Usefulls ..starting with the Nomad 23, a standard 23 kg checked suitcase that physically converts into a twin-door wardrobe cabinet. When you take the removable wheels off, circular bumps on the base seat into grooves on top of another case, letting you stack them into a full-size modular wardrobe.
We’d love your feedback on the concept, the physical design, or how we're approaching modular furniture/luggage!
Just an intuitive and fast time tracking and invoicing app built for people and teams like me who want a quick and reliable way of invoicing clients without paying a premium or suffer through bad UI/UX.
I'm working on Hale, an abstract distributed systems language.
It's a traditional systems language in the sense that you have fine control over how resources are used, but it's abstract in the sense that you have to come to an agreement with the compiler about your intent around the resources, rather than directly declaring the machinery.
It's a distributed language in the sense that there is a higher level structural abstraction in the language that situates computations of the program within a recursive graph structure. Part of that structure incorporates message broker semantics. This gives the compiler a unified view of programs across arbitrary process boundaries. The structural abstractions also enable an effects+formal verification syntax within the language, which allows programmers to codify architectural decisions that are then checked by the compiler.
The compiler itself is infrastructure aware, and the highest layer of functionality it handles is configuring the environment into which applications deploy. The single language handles everything from individual memory allocations to distributed fleet deployments.
The language is close to v1, and I'm currently building a fully distributed inference-as-a-service application with it. Arbitrary hardware/supply matched against arbitrary callers/demand in real time.
I'm working on a diffusion inference engine. I dislike node editors like ComfyUI and wanted a decent python API to control it. The first version was another python/PyTorch variant which was decent but still way too slow to my liking.
I ended up rewriting all the inference in C++/CUDA and I'm finally getting close to limits on the devices I tested. The benchmarks are still a work in progress, I plan to add high resolution comparisons (1536^2 for images, and 1344x768 for video) to give insight in how well this scales.
I've been disliking the recent trend of LLMs releasing with very promising and high benchmarks, only to disappoint by failing simple tasks or sanity checks. Plus, benchmarks don't really tell you about the LLMs' stylistic choices.
So I made 'Model Style not-a-Bench' to be able to quickly eyeball various LLMs, most preferably at 4-bit quants, under 100GB weights (excluding storage offloads like Qwen Flash Next's approach).
We're going to open source everything. Software, SaaS, infrastructure, weights, platforms, protocols.
We're going to bridge local and cloud compute in a seamless way.
We're building an app factory for high throughout development. Reverse engineering tools for non-technical people. We're going to make high quality libraries that others can remix.
We're going to make all of this agentic and AI/CLI first class.
We're going to open source everything and make the world hacker-first again, like Web 1.0 with P2P and RSS and APIs and IRC.
It was born from the fact that whenever I prep for system design I spend a lot of time revising and learning material (as well as drawing from my experience) but whenever it came to the interview I just seemed to struggle to put that together to design the system I was given.
The cool thing about that game is that it uses a realtime environment mesh reconstruction algorithm that allows me to create an approximation of the world 3D geometry and project a dummy “flashlight” with hand tracking. I think I’ve only seen one other game do that technique on quest (some laser tag game).
It runs at 60fps with a tight GPU budget. Consequently it’s not the most beautiful game, but it’s actually pretty fun, immersive and can scale to all sorts of environments.
I’ll be porting it to iOS next (handheld AR style).
I'm making a replacement motherboard for the Nokia 1100 that runs Linux and supports LTE. I have the software working well on a MangoPi dev board and I've designed two PCBs: one "test fit" board that just breaks out the phone keypad/LCD to pin headers, and the actual board with everything on it. Now I just need to order the PCBA which is super expensive :)
It uses an allwinner T113-S3 SoC and the Quectel EC25 LTE module, and I've made sure voice calling works over LTE, at least on T-Mobile in the US. (this is tricky bc of carrier rules)
I wanted the challenge of trying to fit all the components into the small board area available (this phone is tiny), and I liked the idea of creating an updated version of the best-selling cell phone of all time. I figured some of the "dumbphone" crowd might be into it too, if my design works I might make them available for sale
I've built two directory app websites as a fun way to track some topics that I’ve been interest in.
One is https://bannedbookshelf.com which lists books that have been banned for whatever reason, or have been yanked off Amazon for being too ‘spicy’.
The other one is https://fangamesdb.com which tracks the decompilation and recompilation scene which is exploding now with AI reverse engineering tools.
For the past year and half or so, I have been building https://rationalgrid.ai with the goal of bringing together a multi-player LLM interface (based on a underlying graph data structure) and publishing platform, for people to work together to explore deep questions.
It has been tough going, and not really found an audience yet, but would appreciate any thoughts / criticism's.
(Built using Elixir / Phoenix / Live-view / Cytoscape - an amazing stack for web / AI)
WhatsApp released the Agent API but it needs a constant poller to run which sucks as all my apps are serverless that scale down to zero when there's no traffic. So I built Unpoll (unpoll.cc) that takes care of polling, cursor, webhooks and retries. You can simply add your agent's token which is stored encrypted and then send & receive messages by plain old HTTP requests and webhooks.
Lemme know what you guys think and happy to provide a free month to you as beta testers!
Two of my flagship projects I've been working on lately:
Video Commander - A desktop application suite of video encoding, inspection and quality analysis tools (ffmpeg, vmaf, mp4box). I now have quite a bit of paying customers!
Cargocat - Also a desktop S3 client that I created out of some of the components I've done on Video Commander. Newly launched, so still slowly gaining some traction.
Working on / dog fooding my graph based recipe app[0].
There are more things I'd like to do with the graph concept, like linking two user's recipe trees if they were imported via the same URL. Really trying to to avoid features that I think would be cool that don't provide any functionality.
There are still a few features I'd like to implement that are more in the vein of nice to have in any recipe application, like nested recipes, temperature conversions, volumetric ingredients amounts to mass conversions, etc.
But I've been using it for new recipes and converting my old recipes I have written down in the margins to this format and I've been happy with how usable its been in these early stages.
A small tema and I are currently building https://scenema.ai, a platform for creating whiteboard and 2D animated explainers. We initially started Scenema to deal with character and voice consistency in cinematic AI videos. Most video models can only generate 5-10 second clips. More recently that has gone up to 15s. If you wanted to create any meaningful video longer than that, it becomes incredibly difficult to manage your characters and voices.
We ended up moving away from cinematic films and started focusing on explainers for two reasons: 1) we consume a lot of explainers and wanted to produce our own, and 2) traditional animated explainers are very expensive to produce and time consuming.
It took us a few months to get here but we’re now able to generate minutes long explainers ( up to 30 minutes) with studio-level animation across whiteboard and 2d. Here are some examples:
I got tired of always having to update my wife, and my family members résumés. So I created https://theresumechef.com
It's private,free to use, no pay walls or monthly subscriptions.It has10 resume templates to choose from to get you started.The final resume can be exported as a PDF, or stored on Google Drive or Dropbox in JSON format. There are still a few issues, but I've learned a ton by just creating and launching this project. I have some exciting new product ideas, but I'm currently waiting on my Dutch KvK number to accept online payments with Stripe.
I'm working on an IoT automation platform built on top of AtomVM that lets AI agents control a fleet of devices, including device management, telemetry, OTA updates, etc.
The idea is to expose device capabilities (sensors, actuators, cameras) as tools via MCP, effectively giving AI agents the ability to interact with the physical world.
Currently, the platform supports all flavours of the ESP32 and has partial support for the RPi Pico 2 W.
The backend is Elixir/Phoenix, the firmware is Gleam with occasional Erlang.
I’m building a voice-first second brain for Apple Watch. The idea is simple: capture thoughts instantly from your wrist using the Ultra’s Action button or Siri keywords, without needing your iPhone nearby (using cellular or wi-fi).
Audio is transcribed using EU-based speech-to-text with zero data retention, then turned into searchable notes and summaries.
Basically, I want to capture a thought before I forget it, without pulling out my phone.
If anyone’s interested, I might consider publishing on the app store.
For now, the action button on my watch ultra triggers a shortcut the app offers. The shortcut launches the app and starts recording immediately. Takes less than a second. When the action button is pushed while recording, it stops the recording. Quite smooth workflow.
I’m trying to make Siri to trigger the same shortcut with just a voice command but somehow Siri doesn’t do it so that part is still work in progress
Hmm. A lot of smaller projects that seem to be slowly integrated into agent skills. I am admittedly moving into agent world very slowly.
Though now, with the specter of raised prices on the horizon, the local only portion of my setup gained some additional focus.
Still, from the fun mini-projects:
- attempting to replicate testing whether agents elevate inferred persona, memory, or background context above explicit user requests ( https://bloomberg.com/news/newsletters/2026-10-07/study-clau... )
- a few hackaton entries
- my llm rpg got some few additions ( still nothing I am willing to publish )
- first working and useful unraid app;p
I'm currently building AssetJay, an asset management tools for companies without IT departments. When I grew my startup to 25 people, I always tried and failed to keep track of all the devices and peripherals we owned, so built https://assetjay.com to make it easy to do.
This one I (mostly) did without LLMs, but the other project is vibe coded, but I'm enjoying it. An RTS made by Claude in Unity: https://github.com/SiGe11/StarForge2
I gave Claude Code a live stream and a voice. It already build 1700+ games and apps live based on suggestions from chat. It also controls its own appearance and the website itself.
My favorite creation is Heidelberg Blocks. What started as a 3d bridge is now a 3d world with multiple German cities, a karting track, a paragliding hill, a flight simulator and much more.
An app-only lunch bowl chain that can execute very complex customisation without breaking. We've built "cook augmentation" that takes over decision making and QA, added an app-only ordering (accessible to the agents) and ended up with a frictionless experience where you just tell your agent "I'm hungry" and we cook a fresh, hot, delicious bowl that's good for you.
Currently at prototype stage, raising angel round in order to do a live pop-up.
I am building a work/coding agent with my consulting client(s) in mind. They are a non profit and have a software (that I build) to manage property tax exemption compliance.
They have 70+ property managers as clients in the states of NC/SC. A few of the property managers are quite large, managing 10000+ units. The property managers use a mix of Entrata, Yardi, custom Excel to manage the source data. Only a part of that data is needed for tax compliance (which is what I build).
I provide help onboard the property managers. I do all my work remotely, living in India. Lately, my client (the non profit) has been creating mocks of new feature ideas and sending them to me. They use Claude desktop I think. They also maintain a Google Sheet with bugs or new feature requests. I manage the entire development and support via Claude Code.
My idea is to give my client a desktop app to directly manage certain aspects of the software. For example, we support many different Excel/CSV formats for the incoming data. These are exports from Entrata, Yardi or manually managed Excel. Our expected format is fixed but on many occasions there is data missing or poorly managed upstream. My client can easily figure out what is wrong if they get access to the files and generate tasks for coding agent which I can review. Or, if there is data missing, they can also send emails to the property managers. All property managers (my client's clients) could also be given a desktop/web app to help managing the data at source.
I would host a backend for the work/coding agent on a server. The desktop/web app would manage users, permissions and tasks between users and agents. Hopefully, this would grow into a much larger AI enabled product for property managers in a couple years.
Lipik, a free open-source dictation app for Mac (https://github.com/Gourav2609/lipik). Hold fn, talk, let go, and clean text lands in any app: ums and self-corrections removed, spoken punctuation turned into real punctuation. Speech-to-text and clean-up both run on the Mac. On my 106 English test clips, lines that came out fully right went from 28% to 76% with clean-up on. Right now I'm working on Hinglish (Hindi + English in Roman letters). It gets 7 of 20 test lines fully right, so there's a long way to go.
Independent Vet - a website and app that helps pet parents find locally owned (and corporate/PE owned) veterinary practices nearby. It came out of a bad experience we had with our dog 3 years ago at a practice that had been recently acquired by private equity. The sites now get about 30k visitors a day and continuing to build in the evenings and weekends with Claude/Codex has been a lot of fun.
As many of you, I have a few subscriptions to Claude and Codex (just me??). It's a daily occurrence to use 2-3 accounts for the same conversation (whenever the limit stops me).
What I need is something to switch the account where the usage is taken from, not the configuration.
I'm not sure providers would shoot themselves in the foot by not allowing you to buy more usage. It seems like both OAI and Anthropic could use all the subs they can get.
I continue to work on Akariq my travel eSIM business [0]. No app or account required. I have expanded my offering to 200+ countries. In the next 1 week to so prices will come down further in Europe and Caribbean regions.
A lot of people from HN have bought travel eSIMs. I’ve received messages of support plus most people have switched over from Airalo which is expensive and generally doesn’t connect to the best network in the country.
Camera Nocta. I started making an Android app to take better black and white photos of my wife, and it snowballed into a full fledged film-styles app for both Android and iPhone (if I ever get app store approval). It's best for portraits (tuned for a lot of skin) and fashion photography.
The RAW pipeline is my own, with my own denoise, because I hate the "skin waxing" that bad noise reduction and AI seem to do. So there's lots of natural grain. The UI is somewhat opinionated because geared towards more shooting and less editing. The editing tools are my own ideas regarding how to push color (more like dye and pigment than like saturation tools).
It's out on Android, but the Test Flight has more users (1,478) than the released Android app I think. Every single artist that showed interest online used an iPhone, so I guess I should have expected that.
Personal project - long past due to set up a home lab so I'm piecing together a SAN and some compute from old parts i have laying around and some second hand servers in a new rack I put together, along with setting up a bespoke home network instead of just using ISP stuff.
I have long been averse to configuration and customization but my needs and preferences have outgrown commodity systems and default setups, and AI is making it much easier to customize systems, so I'm embarking on this journey after over 35 years of using computers.
A friend needed to decide which cybersecurity conferences are worth going to, so I started building Find a Conference (https://app.findaconference.net).
I'm more of a UX designer / frontend dev, and I've had to learn a lot about things like postgres queues, auth and CI/CD in the process of making it. It's been a challenge trying to develop the features, while also maintaining the conference list and verifying all the info (I use some AI in the data sourcing, but want to keep it all human validated). I'd say it's just reached MVP, and it's definitely still a bit rough around the edges, but I'm already seeing some organic traffic. I'm also learning some conferences are very expensive...
Open to any feedback, especially from folks in the infosec industry!
I’m seeing if an agent can run a full e-commerce website. Nothing fancy, just a storefront that connects to a print on demand supplier. The agent releases new products and designs and makes sure there’s no hiccups in the orders.
I am working on a casual game / experience about Indian festivals as a hobby project. This started as a Diwali night. You can walk through an old neighbourhood down to a river, light 108 oil lamps, set of crackers and send lanterns to a sky that is shared by everyone. It runs in browser without login or any ads.
I have recently added more features like a lamp you keep burning (as per Indian tradition) & a rooftop garden that keeps growoing. I have ideas to add more missions - planning to continue building on this over the year.
Some stats: Around 1,100 players have lit all 108 oil lamps (diyas) & 1600 lanterns released into the sky. Yesterday was the busiest day (because it was Diwali festival)
It's a site for couples to create a wedding invitation website where guests can RSVP right on the invitation. You get a personalized link for an invitation with different domain options.
I’m refactoring https://rcarmo.github.io/projects/piclaw around pi-durable (which is essentially removing overhead and duplicate code at this point, since it was already conceptually close) and wrapping Obsidian into a server-side app with a pi-durable agent to help tag and organize notes. I also have a few Go ports of things like pi-ai and core SIMD inference stuff going, slowly, since it’s too easy to get sidetracked.
I've been laid off in December, and this year I've been trying to get developers off Gitkraken and Termius subscriptions by building 2 Rust native interconnected apps.
I've been working on trying to be more helpful around the house, and trying to spend more time with my kids. I'm trying to give my partner a little bit of relief so that they can pursue their passions as well.
I've been using the recent MIT weight-space programming techniques to work with neural quantum states, which are neural networks that represent a quantum wavefunction. I was able to produce a bunch of novel networks for a few hundred dollars of cloud GPU, which would have probably cost tens of thousands to train in the traditional manner.
I'm really pleased with how my equivalent to Smart Objects turned out in Mojave Studio, my Mac image editor. In Photoshop you can transform an opaque sub-document with an affine and/or homographic transform, while in Mojave Studio you get what I call a "Smart Folder" shown inline, as if you're examining shadow DOM in web inspector. The thing you're viewing has its identity transform and the rest of the document gets the inverse transform applied, to see how it sits in context. https://mojavepaint.app/blog/smart-folders
Late the bandwagon but have been experimenting with agent assisted lean/formalization for the first time; it's been surprisingly high ROI vs previous fuzz coverage and has forced me to think through some interactions of cases that I hadn't yet encountered in the wild. Very useful for flushing out null behavior in my typed SQL engine [1].
On another project to inventory street trees - making progress on training a vision model for aerial tree photo detection in urban environments (original goal was to fill in central park in NYC, which is missing from their public inventory). Haven't gotten NYC to the level of precision I want, but SF has good initial results! [2]
final polishing of FinditAgent, a Windows tool which allows AI agents to research large collections of local files without vector database (more like grep on steroids, for almost any file-format) - and without giving the AI any access to local file systems.
The user may ask any question, the AI will create the search strategy and queries, FinditAgent will act as a kind of Archivar, doing the searches and sending only results or utf8-excerpts to the AI. The tool offers predefined connections to the api of most cloud models - or to local running AI via ollama.
I have developed the underlying search engine for more than 30 years, now added the agent integration. It's now available on GitHub, free for private and experimental use. At this point I'm mainly polishing the documentation and testing how reliably it works on archives other than my own.
It's a product design platform where humans and AI agents can work on the same project.
The main idea is to have a platform-independent system of record for your product that agents can interact with and use to generate code for any platform.
With agents generating more and more code, I think it's important to have a place where the actual product decisions live. Not just the design, but also accessibility, translations, behavior and so on.
You can use an agent to build small apps for you, your friends, family, etc. and use them on the platform. You can also download them and take it with you so there’s no lock-in or special sauce. Its deny-by-default security policy helps you wrangle untrusted AI code and I’m iterating on API connectors to bring your data in. This week specifically, I’ve been working on adding some UX polish to it, figuring out how I can run a reliable hosted version, but you can self host it today. I’m not a designer, so trying to learn a lot and leverage any tools I can!
I've always been interested in how "AI Assistants" work. There's now a ton of them out there. But you can run mine locally, and it has a couple of easter eggs ;-) as well as voice notes through telegram, connect MCP servers, and runs in a single Docker container needing less than 100MB of memory. Any issues/feature requests would be welcome:
Just like the last time I posted in this thread, I’m working on The Marvelous Brass Chessplaying Automaton, a chess database/playing/opening training app for Apple platforms:
The original goal was an alternative to ChessBase, but with an Apple-native GUI and iCloud sync between devices, and as an improved alternative to tools like chessbook.com for operating repertoire management and spaced repetition opening memorisation. In recent months I have focused more on the chess-playing parts of the app, trying to implement human-like computer opponents at all levels, using a transformer-based model (similar to the one described in the Maia 3 paper) combined with a very small search.
Recently I’ve also been working on chess variants. Since a while, the companion Apple Watch app plays 5x5 chess. More recently, I added a 10x10 chess as an Easter egg game inside the app, and I’m now working on hexagonal chess. With the current state of the art in computer chess, it’s almost trivial to implement vastly superhuman strength in all these variants, but implementing weaker levels that are enjoyable opponents is interesting and challenging. Unlike in normal chess, there are no big databases of human games to train on.
I'm working on the interview portion of my federal income tax formalization project. I've hit my coverage target (80% of filers), so I'm doing some refactoring to optimize future development.
https://github.com/emmilco/us-tax-model
Privately I've been running a "game" for the past couple months with claude where it asks me questions about movies I've seen, tries to predict my answers, I grade the answers, and then it builds a theory of my movie taste. It's gotten really good and has reached the point where it can reliably recommend movies and shows I actually enjoy.
I used to watch K-drama on Netflix and pause it to copy paste subtitles in AI chat to do vocabulary breakdown, translation and grammar. So I automated it in an app which auto pauses video when it detects a sentence and gives translation, vocabulary (for every word) and grammar.
I also added same to voice from microphone, for deliberate practice speaking what was learned. To help with spaced repetition.
For about 10 months or so I've been working on this FOSS hobby project to get more practice at Linux packaging, cargo utilities, and migrating to Nix and NixOS (as well as solve some personal pain points in using a HDMI TVs and projectors with PCs).
> TisVlay brings TV Display control for Linux, including: power, volume, backlight, AV mute, input switching plus more, with per-device profile automation via desktop session integration (idle/lock, boot/suspend, remote login/streaming, etc), and in reverse, letting the TV lock, sleep or wake the PC and its remote drive the desktop, while supporting various vendor interface protocols for multi-display setups, mitigating energy usage and OLED panel burn-in or projector bulb burn-out.
Disclosure: this project includes AI assisted coding, documentation and research, and has been a great help given physical disabilities in typing. After programming by hand well over a decade, using voice dictation scene the '90s (IBM ViaVoice or Dragon NaturallySpeaking anyone?), and worsening hands over after my CSE doctorate, this was also a bit of an individual research experiment to explore how effective LLMs could augment my personal approach to spec & test driven development.
While still beta (pre v1.0), its already daily drivable and fully documented. Being largely feature-complete, my current focus has shifted towards distro packaging, and custom ATmega firmware in Rust for HDMI CEC USB adapters to solve a few TV firmware bugs. If you'd like to give it a try, the installation guide and docs are linked below. Or if not, and you already use something else, I'd like to give a shout-out to netflector, especially for privacy-conscious home-automation users, given how common spyware is in smart TVs and projectors these days. The docs cover using it to keep displays on an isolated network.
I am trying to build a better place for people and companies in tech to find jobs and talent. Recently I redesigned the whole platform but everything is pretty much work in progress.
https://uruky.com is a private, paid, ad-free and personally customizable search engine. It was launched by my wife and I earlier this year and reached 400 MAUs and €2000 MRR last month.
https://podby.com is a simple EU static and app hosting service. It was launched last week, and is at €20 MRR.
https://fairats.com is a simple and fair Applicant Tracking System for remote-first teams, launched a couple of months ago, still at €0 MRR.
Happy to answer any questions and consider any suggestions!
Thank you! It depends on what you care about, because it has less features (it's not trying to be an ecosystem with Mail, Browser, etc., and it doesn't have AI features, for example), but you can control which search providers to use and you get access to the source code after paying for 12 months, which Kagi doesn't do.
Working on Aqueous [1], a next-generation compressed archive format. I've just finished a long refactoring session that enables streamed processing instead of keeping intermediate results into the memory or spill file, and am now working on formats that were big enough that required this support (e.g. disk images). Meanwhile sparse file handling and in-place move and delete operation of archive members have been implemented.
Its slow going. When I started this near the beginning of the year the ratio between input and cached tokens costs for the interesting open weight models was less than 10 and provider side cache hit rates weren't very good, so it wasn't expensive to break the prefix to remove old things from the context and do rolling summarization to keep the prompt size down.
Now some DS v4.1 Flash providers have a ratio of 33 to 50 and >99% cache hit rates. That makes the cost of prefix cache breaks a lot higher and its definitely harder to get ahead on total cost* of the usual approaches to context management for DS v4.1 Flash specifically.
* - the number of requests is reduced by a factor of ~3 relative to oh-my-pi on the same benchmark; the task is long enough to require compaction once when using oh-my-pi
I needed Postgres backups at work and at home at the same time. I tried a few existing images, but after spending time setting them up I found they shipped an older pg_dump that couldn't back up newer Postgres versions. So I built my own.
It runs pg_dumpall on a cron schedule and streams it straight into a restic repo, so nothing touches local disk. You get encrypted, deduplicated snapshots (hourly dumps of a mostly unchanged DB take very little extra space), retention policies, and daily pruning. If the dump fails, no snapshot is saved. It ships pg_dumpall 18, so it can back up anything from 9.2 to 18.
I now run it in production at work, so I'll be keeping it up to date. I'm no expert, and this is the first time I'm doing something like this (putting something out there), so any feedback is welcome, especially from the database pros.
It's not fancy, but it's something I needed, and boy, I really needed something like this.
Thank you! I honestly wasn't even aware of rest-server. However, I just tested it against rest-server and it works.
Set RESTIC_REPOSITORY=rest:http://your-server:8000/your-repo, put your credentials in RESTIC_REST_USERNAME/RESTIC_REST_PASSWORD (or in the URL), and leave out the AWS variables.
So rest-server seems to have an --append-only mode. If that's enabled, backups still succeed, but the retention step is refused, so each run reports an error and old snapshots are never removed.
I'm building Starkeep, an open-source serverless self-hosting system that lets you keep your own data and build your own apps that run on your devices and your own AWS account.
Starkeep's common core provides file indexing, syncing, access control and more, making apps both easy to create and interoperable. Starkeep is easy to deploy, and its fully serverless architecture minimizes running costs. I'm working on an official Starkeep Photos app as well (https://github.com/starkeep-org/starkeep-apps), built for web and Android so far. Everything is MIT licensed.
The first prototype PCBs for my automotive grade, TockOS (Rust) running controller are on order. I've been working on it for a long time, and I'm curious what sort of traps I've let AI run me into. At least for the layout I had to re-do it three times and continuously push the AI to make certain considerations.
Once the design is finalized, I'm wondering what price point I'll actually be able to hit. I originally wanted to be in the $200-$300 price point to be similar to [rusEFI](https://rusefi.com) but I think it'll likely land more around $400. Note that my controller isn't for engine control, just a generic control module, but it DOES actually use AEC qualified parts.
After I'm done with this design I was wondering if I could make a RISC V off the shelf engine control unit, but I'm trying to keep my focus on the TockOS board (Chameleon) first.
working on AudioWeave, an ad free, offline first Android podcast player.
switching from iOS, I wasn't happy with the existing podcast apps, so I built this mostly for myself. I've been using it for the last six months and finally pushed a public beta.
My day to day is BE so I used the project to learn Kotlin and Jetpack Compose. Everything is strictly native using Room and Media3/ExoPlayer. I wanted a nice looking (subjective), privacy-focused app without any tracking or ads.
Few apps I have originally mainly done for myself; Vaava (https://vaava.app) , been tweaking the e2ee data sync between connected apps, and the webrtc based baby monitor feature stability. Crazy how easily Astra / opus 55 level models create quite complex ad-hoc reproducible (almost automatable) test scenarios with 2 phones, laptop and other tools / peripherals they know they can use, for example for the baby monitor feature Astra did a setup for me which i can reproduce to test that the video + audio works fine between 2 phones for x time.
Also trying to replace own Duolingo usage with Fika Lingo (https://fikalingo.com) basic idea was to first repro the practice of matching words. There is now few other types of language / vocabulary practices. The practice generation works on-device with local LLM, you can give the topic yourself like "Words that relate to Hacker News". Downside at the moment is that English needs to be the counterpart of the practices. Cant guarantee good enough quality with something like germany <-> italy practices. I'm trying to do finetuning with LFM models to see if it would be feasible to improve the local practice generation.
Should I write about the work?
I create agent routing. The key is how you define an agent. I define it as a package of memory, skills, and its own scripting tools. Within the deterministic code, it's not the agent that operates but the automation. Assigning responsibility and receiving probabilistic responses is the same as using LLM. However, as you go through life, agents gradually accumulate, eventually reaching dozens or hundreds. It's a waste to include all of this in the system prompt. With our self-developed routing flow, you can automatically select and use only the ones that perfectly match your task.
I'm working on Eigakan Guide (https://eigakanguide.jp/en), a cinema showtime aggregator for Japan. Every popular cinema site in Japan only allows for very few filters, making it hard to find exactly what I was looking for. So instead I built a site where it's possible to filter by area, cinema, date, time, features like IMAX, English audio, English subtitles, and more. This allows for searching for stuff like IMAX showtimes with English audio on Friday after 7PM.
Since the last time I posted on HN I've added 100s of cinemas, user features like being able to save search filters, movies, and specific showtimes. Also a ChatGPT plugin which allows for searching for movie showtimes using natural language directly inside of ChatGPT (https://chatgpt.com/plugins/plugin_asdk_app_6abfdc298c208191...).
Finally I've added AI search to the site as well, allowing for natural language search with queries like "English audio showtimes near Shinjuku tomorrow" and so on.
I would really appreciate feedback for things to add, things that are hard to understand, and anything else!
I'm not a code guy. I work in medicine. I've been working on a tool that can automate part of my work. If it works well it can free doctor working time and save lives too. It's related to ECG interpretation.
A personal AI librarian that makes large collections of local epub's and pdf's accessible again. It helps people rediscover forgotten files, find information by meaning, explore unexpected connections, and save discoveries for future interests or projects.
Now I'm going to be working on my Linux-subset ABI x86-64 kernel [0], specifically adding support for tracing, then working on allocators. It's written in Rust so I might have to figure out how to make allocation fallible like the Linux kernel does. Also this raises questions about approaches to overcommitting and OOM handling, which I feel Linux does badly, but when constrained to Linux syscall ABI there doesn't seem like many good options.
Also I am not using LLMs for this since this is a learning project, with near-zero real world utility -- I just want to understand OS kernels as best as possible.
All Voice-to-text pipelines for linux sucked, so I made my own. I also patched upstream libraries like ONNX to actually work with latest models (was a PITA to make parakeet work!)
It's now the fastest and most accurate linux transcriber.
No, you don't! It's fully automatic training based on your local voice clips. Nightly or Weekly, when your PC is idle, it uses a big fat model to transcribe your clips, learn, and train the faster smaller everyday model.
You CAN do manual training and corrections; it's helpful, and the hooks are there, but totally unnecessary. I don't do it, it's good enough as is
Working on ZShot [1]. I'm about a year in on a local-only browser tuned specifically for CLI and API usage - kinda filling the gap for "geez, I sure wish I had a browser, but not one to look at". You need a webpage as PNG, or an MP4, or a website as a WACZ or SQLite database. There's a Mac app to help with discoverability of options, but I wanted for my local tooling and building services on top of it.
Along the way I ended up adding an internal agent framework to satisfy stuff like "log in, click the button, and then take a screenshot", out of which fell stuff like automatically generating Playwright tests from that. It even runs the Playwright test suite and Puppeteer test suites ( a bit faster, and for a lot less memory ). I kinda love this bit because I can layer in multiple models that will bounce control of the web session between each other, to do stuff like let a Typesafe model choose to escalate to Opus when it needs higher minded behavior. It's fun to watch it happen.
I've mentioned it a few times around here, I'll probably do a Show HN soon when I feel like it's ready for that. Hopefully soon.
Thesoapbox.email - a social experiment email lottery to see what people do when suddenly given enormous reach. Every day a new subscriber is drawn to email the whole list. It’s hit 10,000 views in a couple of weeks which is my most viral project so far.
The second - a personal AI learning platform built on recursive self improvement and tailored content to push my boundaries. I’m a lifelong learner and this unifies everything for me.
(Currently not public)
Continuing with work on my weekend hobby project https://colorhunt.lol You get assigned a random colour and have 24 hours to go out and take photos matching it. The game then generates a photo grid from everything you captured.
Modes:
- Solo → 9 photos by yourself(added 2 new modes, a palette where you are assigned 3 colours and a rainbow where you get 9 colours)
- 1v1 → compete against a friend, combined grid at the end
- Squad → everyone contributes to a 20-photo grid(added a room option where you can take a break, react to photos uploaded or just join as a spectator)
No accounts, no app install, no personal data stored. Photos and generated grids auto-delete after 24 hours from Cloudflare R2 storage.
Too bad someone stole my MacBook just after adding the palette mode. Waiting for M6 to start selling then resume dev work once I set up the new laptop.
An opensource personal expense tracker, I was a happy user from an offline expense tracker which wiped my data when playstore decided auto-updating it,
I'm at the point where I'm finally actively using it to register my activity, it is straightforward to register tickets by sending them to a telegram bot, then, reconcile those by uploading my bank statement.
I'm trying to get the lowest possible usage friction so that I keep using it.
So far the better built pieces related to the data ingestion. Once this is good enough I'll build reports + anything coming to my head as a useful piece.
I'm leveraging AWS textract + bedrock + bda to parse bank statements. I expected this would have been easier but it is still a very complex problem which had required me to put many iterations (so far it works nicely with my own statements)
For ticket parsing, AWS textract + parsing rules has been enough.
It's a cloud where you get completely persistent Linux computers that share a group of resources (CPU, RAM, Storage). This means that you're able to allocate resources across workloads which has a ton of benefits when working with agents.
I found that current cloud infra was extremely restrictive with the new tasks & ways of working that coding agents open. AI is going to continue to change the way software is being built, and I believe that the infra we work with should also be malleable enough to give AI freedom around computers and resources.
The main principle I've had behind it is transmutability -- so computers and resources can be shaped to match any workload.
A friend of mine recently expressed interest in writing a story with me, which gave me the drive I needed to pick back up a project I abandoned a few years back. It's a website for writing a story one block at a time, branching off wherever you feel like you'd want to move the story. I've also used it as an excuse to play with Claude a bit and fiddle with ideas.
Away from the keyboard, I've picked up acrylic painting. I signed up for classes and have been enamored with it, so I purchased the most basic of supplies to paint at home. It's been an absolute blast, and there's a sort of thrill to being bad at something and still finding joy in it.
I know how hard it is to distribute your product and find developers at scale.
So I've been busy creating a solution for early stage companies who are low and budget but still want get in front of developers and increase awareness to their product. Hopefully it can help others.
daily.dev for Startups:
https://daily.dev/startups/
Allegedly, token prices may go up soon, so I'm trying to use tokens more efficiently on my $100 sub.
Noticed that browser agents are not efficient with tokens and also kinda slow. I’m familiat with the web, so I made https://bopbop.dev to control browser and do same things faster and with less tokens.
It is a Chrome extension that lets Claude Code, Codex or other agents use your browser. Comes with a skill/mcp to connect to it.
I used it every day for a month. Now I use it instead of Claude in Chrome.
I made it public last week so it seems fresh but I’ve been cooking for some time
I have been playing with benchmarking agents in a research loop. The experiment leans on using these models to understand and explore builds/gameplay through reasoning game AI over traditional reinforcement learning.
Also, creating interactive ways to view the experiment and visualizating the results.
A data-flow-based, lightweight formal verification engine for application security. Basically, it's SAST++:
1. Deep enough that you really have some formal guarantees about the codebase (like zero findings actually means something, not just that the tool is broken).
2. Fast enough that you can shift the formal verification phase left and apply it to every feature during development (for example, during agentic development sessions to get fixes done within the same feature context).
3. Expressive and customisable enough that you can write specs for most vulnerability types from the OWASP Top 10 (even Broken Access Control and business logic flaws). Skills for writing specs included.
the plot: Programming languages existed as an abstraction for humans. As writing code becomes cheaper, we can write it for each architecture as long as the specification is done well. The costs of running will matter more in relation. So don’t make your AI write rust, but assembly
I've been building https://botlark.com/, which is sort of a HN for autonomous agents. The skill is on the site and on ClawHub if you'd like to take a look. Feedback very encouraged.
AI-first cookieless EU-hosted web analytics. Give your agents access to your site's traffic, client & server errors, and page speed. One script tag install, no cookie banner, free trial.
Started off as a side project monitoring traffic across a network of websites & API/MCP endpoints. We run our own custom CDN ensuring no data at rest in the US or at a US company.
You can always see my current projects on my GitHub profile README, I’m pretty proud of it and get great feedback on it regularly. I like to think I don’t build standard crap.
I’m also working on an interactive terminal workflow tool, it’s not quite ready for the world.
It’s effectively an interactive command cheat sheet with its own built in shell, and compostable workflow recipes, all backed by markdown. So your commands and recipes can be read by humans, and executed by anyone (or anything).
Warnings or Cautions in markdown become warnings and cautions when running commands in the shell.
1-year part-time experiment: can you build a large project with AI writing the code and AI plus deterministic tools verifying/guiding/enforcing it for structural and behavior consistency, without a human reading the code?
I'm building a polyglot web framework in 15 languages, with C# and Lean4 as the reference implementations. Structural and behavioral audits using deterministic tools check that every port stays consistent with the reference. It's about 1.8M lines now. I have not reviewed the code/neither have i verified the behavior tests except at the time of giving requirement and never seen some of the language code at all.
The question is how far can we produce "working code → correct code [Possible] → good code [Unknown]" with deterministic audits correcting structure, ai mistakes and behavior validation.
So far the quality looks acceptable with lots of correction of AI output done without looking, but it isn't fully verified.
I use hypertext to map reasoning and arguments. I have AI skills to map out a tree of arguments, all of which are given an initial "debatable" rating on truthfulness. I have a UI in which a human can dig through the tree and mark statements on a 1-5 scale of false/dubious/debatable/plausible/true. Last night I implemented 3 things I identified 5 days previously as potential improvements. https://writing.brlewis.com/posts/2026/my-ma-ballot/
30 years ago I learned to program by building games in Cocoa / Stagecast Creator. I recently partnered with Dave Smith (one of the original authors + members of Apple's Advanced Research group in the 90s, and low-key the inventor of the icon) and we're rebuilding it for the web:
I'm user testing it this fall with game-building jams in local libraries. Obviously you could build games with Claude, but I think programming-by-demonstration is a great way to introduce fundamentals, and the simulation environment has a low floor and a high ceiling for creativity.
A self hosted software forge / factory / agent platform / mini heroku thing. It's not polished yet but it's building and maintaining little apps and games. https://github.com/flokkit/apptown
You can try it out on mac or Linux in a temp container.
It's self hostable and can use local models. It's not intended to ticket crush at massive scale or produce huge software overnight. It's for deploying smallish things in a predictable & consistent way, with low friction. Then help maintain them long term (e.g. take care of dependency and container updates and similar chores).
SuperBlocker is a native blocker that works on iPhone and iPad using Apple's latest URL filtering technology that blocks URLs without revealing it to Apple's relays or SuperBlocker's servers.
The original JSON Query app on the Mac App Store. Store, organize and query all of your JSON with jq or natural language. All of your data stays local so sensitive JSON can stay on your machine.
- hermitator, a tool to run multiple isolated AI agents for WhatsApp on a single machine, with idle agents using no RAM: https://www.encona.com/posts/hermitator/
- showntell, which allows your coding agent to explain things to you with a video that starts playing within a few seconds of you asking: https://www.encona.com/posts/showntell/
- a philosophy tutor (testing in private for now) that does 2:1 live philosophy lessons, each exploring a single topic (aimed at ~10yo kids). I've tested this successfully (i.e. two kids did an hour and they want to do it again), and am trying to make improvements.
Oh, showntell is nice! I'll try incorporating it into my project (obeya.si), which already uses demo videos to show me work it did. But when I ask a question, the response is still just text. Could be so much better with your solution.
A fully FOSS Church management platform called Churchit ( https://Churchit.app ).
Jesus once said "Freely you have received; even so freely give".
Churchit aims to be a comprehensive and feature rich all-in-one solution that churches can self-host with minimal technical expertise.
Churchit provides churches with a dynamic public website, a member portal, and a feature-rich management backend.
a mac app that listens to your microphone and gives feedback on your pace, filler words etc in real time. Built for myself as I tend to get carried away in meetings. Looking forward to feedback if anyone else has the need for something like this.
Just launched https://otusanalytics.com. It's a European Google Analytics alternative with strong privacy options and modern features such as an AI assistant that can act as your web analyst
I've been building an art therapy sandbox upon Pavel Dobryakov's fluid sim, and I'm almost finished, and going to release on Steam, and shortly after, I'll be releasing it on GitHub fully open source.
Over the years I've used it myself and have found it very soothing, slowly adding to what it can do. Hopefully others can find it soothing and beneficial, just as I have.
(semi-related, I'm seeking design / design engineer work, I'd love to be pointed in the right direction if anyone knows of a role that might fit, my email is in my profile)
For the past few weeks, I have spent my hobby as building and improving my own coding harness, VT Code. I plan to keep improving it throughout this month. Recently, I built a native macOS using GPUI (gpui.rs). Docs: https://github.com/vinhnx/VTCode/blob/main/docs/development/.... I'm pretty happy with how it's turning out, though there's still plenty to improve
I’m finally doing the thing I always said one day I would, and writing a game! Early days yet, but I have proof of concept for much of the rendering tech now.
I was working on platform to build 3D room design for everyone who loves interior design, no need to install, no need to pay. You can see my published design here :
https://formanest.app/p/2LxQNrjg50vvxrbb
I'm working on a Markdown editor for writers & researchers (https://grasmere.app). A bit like a cross between Obsidian and Scrivener. I'm finding the marketing part hard but slowly getting there
Also hacking away at a little encrypted messaging / key management app, where all the crypto happens client-side so your private keys and unencrypted message text never leave your device:
I have been in enterprise for couple decades now, learnt and experienced the challenges/gaps between the specs vs implementation, the asks vs the real needs in the space of identity, security and access
Product: webfrolic.com
Demo: demo.webfrolic.com
What it is?
Consider Okta (Workforce identity)+ Auth0 (Customer Identity), Authzed (authorization) + ServiceNow (IT service Desk) + Zendesk (Customer Support Desk) architected together
Being an engineer there is this inherent habit of prioritizing correctness over marketing so I would not oversell it, though its probably the most secure and complete thing that could be built from identity perspective
Status (WIP):
1. I am still debating/settling on how to incorporate the AI Agents needs into the architecture, big focus on avoiding the creation of security or performance foot guns where administrators leave room open for abuse with configuration despite all this attempt to build secure system
Hence (https://hencegame.com) is a modern twist on Einstein's puzzle, with fun picture clues instead of text. It's a free online daily puzzle, the same for everyone, where every puzzle is always solvable by pure deduction with no guessing.
I launched it October 5. The hardest part was the graphics, making each picture clue unique enough that you recognize it immediately and know what category it's from, even at small sizes. Easier said than done.
Every puzzle is run through a verifier first to make sure it is solvable without any guessing, just by pure logic, so there is always a next step to find, which is what makes it feel so rewarding. There's a hint system to help and it works really well if I do say so myself . No account needed to play.
From what I've been hearing so far, the people who have checked it out are enjoying it and can't wait for tomorrow's puzzle, so that's a good sign.
I'd love to hear how the difficulty feels, and whether the pictures read clearly.
I've been working on a multiplayer card/board game with some friends! We've been using Unity (C#) and its services like Relay and Lobby. Working with Unity has been pretty solid so far but I cannot recommend Relay and Lobby (now multiplayer tools) as their docs are missing a lot of crucial details such as how to preform host migrations.
We are hoping people who enjoy Magic/Hearthstone will appreciate the complexity and mind-bending nature of this game:
For now Im working on an agent-orchestration application that is taking tickets from a tracker and giving them to an Agent. It's called Even Keel.
The main feature is, that I can use different subscriptions/API keys (Claude and Codex - v0.1.4) and trackers (GitHub, Linear and Huly - v0.1.4) to make Even Keel start an agent for every single issue, task by task, one at a time per folder. There is also a parallel feature, so I can run more than 1 project at a time.
Once upon a time, when codex agent was integrated into the app, I wrote a ticket to make codex his own hook. It reused a hook that blocked its own git merge, if tests did not pass. I started calling it "A new agent writes its own cage"... Right after, it turned out it doesn't listen to the hook and it fixed it
Version 0.1.4 is already shipped, but v0.2 is coming soon with more supported trackers, agents and new features.
I've been working with orchardist and gardening groups to build an app to plot and track your trees over time. It's useful to be able to look back and see when you graft trees, or when they fruited over time.
I’ve always wanted an interactive history timeline, so I tried to create one [0], but I am not happy with the result. It’s difficult!
[0]: https://jonasfoyn.com/timeline
A slow, realtime strategy game similar to Neptunes Pride and Subterfuge.
Each game tick is an hour and games can last over a week. I got hooked on both Subterfuge and Neptunes Pride and always wished there were some more titles in this genre.
I've been playing bossa nova guitar for many years and I've always wanted to share the chords of some of the songs I've learned by ear in a fun way. Since I have zero front-end skills and little time I was never able to. Now with the help of a model I created a nice webapp that syncs the chords with their youtube videos (the chords were all manually added, only the UI was vibecoded). It's been fun and a great way of learning the names of more complex voicings.
Rebuilding my ERP to work more like the model of https://plaintextaccounting.org, where I expanded the idea of `double entry accounting` to `double entry trade between parties`, plus several other ideas for what I see a way to collapse a lot of complications that are typically for ERP systems.
I make a special `textual DSL` where I model business scenarios like `Make a sale that include 3 invoices : 2 to different cities in country A in currency USD and other in country B in currency EUR` + all the intermediate steps like transportation, manufacturing, etc.
I`m now in the woods with a simplification for the price engine, I dislike much how complicated is normally, and I think I have good enough solution for most cases that generate without dynamic calculations.
Given the popularity of decision models I tried to create a standard text classification pipeline that runs entirely in the browser using webgpu/wasm. Downloads an embedding model, trains the classifier (a Linear SVM for now, whose hyperparams are determined by cross validation over a grid), calibrates the classifier so that reliable confidence scores may be produced. This is in Python that is run using Pyodide. You can paste your own data (no uploads supported) to train a model on it. WIP, and mostly generated using GPT Sol. Testing a few things out. Will have a blog post soon.
I wanted a way to manage multiple loops without babysitting terminal sessions and to talk to them as they work. With Spool, you can create and manage loops from a web UI, chat with them through Telegram or Slack, and let them communicate with each other. Each loop can run on your environment directly having access to everything you have, or in its own container or even better on Anthropic's cloud.
I’m using Spool to build Spool and figuring out what works well when several loops collaborate on the same project. That's also an interesting experiment going on.
I’d also love to hear how others are managing long-running or persistent agents.
I'm volunteering and running meetings - and I wondered how tricky it would actually be to get decent meeting minutes and record a (small) multi-person in-person meeting.
Turns out there's a few tricks to recording a room on Android - and rather tricky to get good meeting notes (in Norwegian) from the audio using only local LLMs - and avoiding the data touching the cloud at any point.
I need to revisit Android recording app recommendations - and what combination of incantations will allow you to transfer an audio file directly to a MacOS laptop (apparently it's not quickshare and airdrop unless your Mac is wide open...).
But I have a wip single download app that lets one record and transcribe - should work on MacOS (tested), Windows (not yet tested), Linux (tested on chromeOS/Terminal).
What might be of interest here is the launch video I made for my app, Fit Shapes—combination of Claude turning the app screens into a video, GPT Image 2.5 for image stills, Fal.ai for animating the stills and Elleven Labs for music. An interesting thing about agents creating music is that they can create and adjust tracks in segments to bring in specific elements as needed.
I know there are a thousand of macro trackers out there, but most of them are some combination of ugly, tedious, expensive. Since it's an app I use every day I tried to make my own with a focus on removing taps.
How to make data storage on multiple storage devices unbreakable and consistent? The former means for me that back-ups should be automatical and serverless, the latter means that no matter how much of that HDD/SDD devices are connected, I want always to know all metadata of all files I have. I want this software under GPLv3 if to consider GPL still protects me from anything because permissive licenses are no better than WTFPL by any single case.
My first idea is to require the user to have more than one disk for the system needs and the system should know all of the rest data devices user owns. My second idea is to treat different files as having different priority which impose multiple copies. My third idea is to make the software accessible, and the last idea is to make encrypting of all the devices easy and with no ties to any commertial entities. I would like to have a discussion about how to not lose anything digital.
It's a bit stupid but I made a free horoscope api during covid and hosted on vercel, it started getting a lot more traction since and I have migrated to a proper domain and to cloudflare workers. Getting around half a million request / day and bill is just below $10
- Scripture: https://scripture.wolftrail.net/ - a website I use to daily read a couple of Bible verses according to the Russian Orthodox tradition. It's still not fully functional, but useful for that purpose at least.
- MacGames: https://macgames.wolftrail.net/ - reverse engineered vintage Macintosh games (System 7 era) that can be played in the browser.
Been working on my version of a journal app that is basically Diarium but I can add any features that are missing from Diarium and Day One. Been a cool side project.
I wanted my journal to have a proper “writing” mode where everything disappears and it’s just the words.
Better voice to text transcriptions. Ability to paste the transcription into the body of the entry but also keep the recording.
Better picture layout.
Apple Watch integration so I can just speak into the watch when I have an idea or something I just want to say.
And when I think of something new it’s very easy to integrate. Love it.
For the past two years I've been working on my day planner app Ember.
Most planners treat your schedule as fixed, so the moment a meeting runs long or a task takes twice as long, the plan is wrong and you either fix it by hand or give up on it. In Ember, if you move a task or change how long it takes, everything after it shifts later, and it flows around fixed things like meetings. It's been super helpful and I have been adding new features ever since.
Right now I'm working on a native Kotlin app to replace the current one is web + Capacitor and exploring full duplex voice models so I can add or move things around the app while having a conversation with an agent.
https://santawho.com A funny holiday game you play in person with the whole family on just one phone (the parent interviews everyone for jobs on the North Pole and surfaces inner insights about what everyone might enjoy from Santa). It’s for any age, private, ads free, just pure holiday spirit.
https://www.dreamlist.com - collaborative shopping lists and online wishlists and gift drive platform. Again ads-free, private, stack multiple lists into groups of lists to share on social media or with grandparents.
A number of other ones about to launch as well. All little tech, not big tech - private by default, no ads, no feeds or media, just surfacing what is needed.
GeoPeek, a small macOS Quick Look extension for GeoJSON. I work with geospatial data and wanted to press Space on a .geojson in Finder and see a map instead of raw JSON. It also shows an attribute table and basic stats. https://github.com/serifsadi/geopeek
VibeVoted (https://vibevoted.com/) - a platform for product assessment and competitions based on the rankings of 10 leading AI models from 10 different labs. Also, each product get's feedback and recommendation for improvements. Here's a sample voting session: https://vibevoted.com/s/password-managers-2026-oct
A GEGL/babl based image raster/vector/pixel editor for Mac. Using metal for all effect on-canvas previews to both have speed and full xcf compatibility.
An ide that has a full agentic plugin that allows ablating functions or pieces of code that one has to complete. So instead of leet code you can build towards a program and still learn something. It can also guide you with bug fixes and other stuff. And it also includes a fully integrated dos environment with a fully functioning debugger and asset editors. Also works on iPad.
I am working on a Surf app for Nordic countries, the wave layers presented by the large forecast groups are too course to estimate good waves based of local winds in very complex terrain. No public demo yet. If you have any experience working with weather apps pls link some ty
I'm getting hit constantly by bots trying to hack my site, so I'm trying to hack my own sites. Pretty much I have a few obliterated LLMs (DeepSeek-V4.1-Flash-Obliterated, OrcaCyber-Zero, etc) which run on their own sandbox to try to hack me.
If anyone wants to try it, you can test it on mockhack.com
Lately we've been working on access for AI agents. A lot of people want their own agent to do the repetitive part of time tracking. Control remains important, which is why we rebuilt our core this month to support change history and full undo/redo functionality. Feel free to try it with your own agent.
We built this after the timesheet tool we paid for raised its prices and moved some features that we used regularly behind more expensive plans. We had to add three new colleagues, and everything combined this would cost us $500 a year more for the same software.
We run Elixir and Phoenix LiveView on a Kubernetes cluster made of Talos instances.
We're still looking for feedback on the product, so please let us know what you think!
We also added a live-demo functionality to make it easier to try. It spins up a workspace just for you, with sample data in it. You can make that workspace your own by adding your own data and undoing the creation of the sample data via "History". It should just take a single click.
Meli: A spending tracker for Mac designed with privacy in mind. Turns credit card alert emails into spending insights and budgets. I was in security for 15+ years and I built it because I didn't want to use Plaid or other aggregators and wanted my financial data locally on my device and not in the cloud. Spreadsheets required a lot of work to maintain so I wanted some sort of automation. Connects directly to your email (reads only alerts from supported senders) and all parsing happens on your device.
Works with seven major US banks (AMEX, Capital One, Chase, Citi, Discover, Bank of America and Wells Fargo) plus PayPal and Venmo.
Does this list cover most of your spending? If not, which banks or cards would you need?
Like 2h ago I was doing some research on that topic. I’ve heard about open banking a while back and thought: Oh that’s easy. I’ll just get an api key from my bank or “oath” into it.
Sadly it’s not that easy.
Unfortunately it's not easy. In the US specifically banks don't generally offer APIs or keys to access them to individuals and you have to go through an aggregator like Plaid (which is what I want to avoid).
However, most banks do send alert emails when something is charged to a credit/debit card. Meli connects to your inbox read-only via OAuth (Google and Microsoft) or IMAP for other providers and reads those emails on your Mac.
Of course there are limitations to this approach, you only get access to the data that is available in those emails like merchant, amount, date and account.
Which banks are you interested in? Are you based in the US?
At work, I'm setting up a harness to make research, experiments and doc work more efficient, and at home, with the goal of working with a game engine, I'm trying out various tools besides ChatGPT and Claude, like music, sound effect, video generation and motion generation. What I feel is that even the companies working on frontier models still don't seem to do well with sound that isn't voice (songs, sound effects) and motion understanding (not video, but things like moving a character) (though they might have it internally and just not be releasing it).
I couldn't find an open-source inventory management system I liked, so I started working on one. It was designed to handle multiple stores, each with their own inventory and products. Each product in categories and sub-categories, and location management for each item. It was supposed to be the base of a few online stores and a brick and mortar shop, but that business plan fell through. Still working on the inventory system, and it still needs more work.
Built an app to suggest best local beach for given weather for my town. My partner shared with local moms and they seem to be loving it - https://beach.metacursion.com/
Also revived old project - https://aminosplore.metacursion.com - the idea was to compare what essential amino acids each food contains using spider web chart. It’s been dead for a while due to USDA API changes so I revived, added macros and enabled to build foods into actual recipes.
Finally have been improving elsrc.com - an easy and scalable way to use ai image generators. If you need hundreds of images you can just request them by url. Supports versioning, batching, regional and national styling based on viewers IP and few other neat features!
My co founder and i (aka Opus 5.5 and me) have been building https://www.dexbindr.com/ for the past few months now. It started with wanting to complete a pokedex binder but i hated carrying my 11 pounder card binder to card shows. Decided to build this app to help with card pricing, scanning and tracking to get my binder to completion. And apparently people actually do find this useful enough to pay for it so another example where solving your own problem can also help others.
The bottleneck in most AI coding workflows is the actual PR review, so this just comments on open and update, including drafts. I tried a bunch of setups until it was cheap enough to leave on and still fast enough that you're not sitting around waiting.
I'm working on an open source agent and engine that can calculate your Federal and California taxes. My goal is to build an AI system that can act as your CPA, and hopefully allow you to do super cheap tax estimates and planning.
I’m building Spine, a platform for agentic applications because I want agentic applications to feel more like traditional software: something you can reason about, inspect, reuse, and distribute.
I try to 'vibe code' a product every 2 weeks. Most of them go unpublished, I just build it for the sake of building it. Most recent project is Unmute Party (https://unmuteparty.com) - multiplayer party games for remote teams, played in the browser during a Zoom or Teams call. The server is EU-only for now, and I plan to add US servers next week. Most of it is free, the paid tiers aren't even working yet. The paid tiers can be unlocked by using code HNWHATAREYOUWORKINGON.
Apart from that, running a travel agency I'd never be able to without agents (they manage my ads, social media, website content etc.). Started with boa.ba and it went quite well, now preparing dobosnia.com
I'm building Rank - a language I could comfortably write on my phone. I'm even thinking about making a dedicated handheld with a physical QWERTY keyboard and multi-function keys like the ZX Spectrum had
FreeBSD 15.2! Scheduled for release in early December. Biggest change in 15.2 from my (release engineering) perspective is that I'm using got instead of git, dramatically reducing the dependency tree.
Column view for hacker news with resizable fonts. It’s just a fun hobby thing for me. I’m not in software, but I can do hobby level programming and that gets leveled up by codex.
Writing procedurally generated civilisation, mostly inspired by Crusaders kings mechanics, with personalities driving most of historical events. Haven't had so much fun in a while
I wrote a children’s story in Galician about open source. It's told as a fable. I made it open source, and people are now helping translate it into Spanish and English.
I’m looking for ways to fund printing and distribute the book. If anyone has done something similar, I’d love to learn from your experience!
Visor. It’s a new type of trading terminal centered around options and gex. The main feature is overlaying the options payout graph right on the chart.
Think tastytrade rotated 90° and put on your live barchart and then watching all the Greeks play out in real time.
Has been instrumental in me deeply understanding the dynamics of more advanced options trades.
I'm working on a table tennis VR game with human-like AI. It's trained on roughly 25 hours of me playing against myself and runs in real time on a Quest 3: https://loopspin.gg
Which let me deploy pi-durable agents to AWS, Cloudflare, Fly.io as services with every cheap cost(in MicroVM)
What I assume that the agents should run as a "Web Service" on the cloud and do the work silently and complete the work without the chat. We don't chat with our system monitoring, don't chat with log system.
FoldRoute - a PWA route planer for folding bikes & public transport
I bought a Brompton a few months ago and realized there is no good planer for combining biking and public transport.
I one-shotted the first approach with codex. It was so decent that I wanted to elaborate on the idea. Still not perfect and needs a lot of testing in the wild (especially the navigation).
Currently, only in German. If there is demand, I will translate to English.
For now only a PWA. Therefore, it has some limitations. Maybe I will make it an Android / iOS app one day.
I am working on an open source and local app for tracking one’s wealth.
I wanted something without ads, subscriptions and cloud, where I could throw in all my accounts, stock and crypto positions, loans, funds, estates etc to have it stored and tracked on one place and via MCP exposed to my Claude financial advisor.
I originally thought it would be useful to have a package registry that helps prevent supply chain attacks. It reads and judges all of the changes to popular NPM packages. While building I realized that it’s a package manager that agents can settle package versions with as old versions have new vulnerabilities found.
I’ve caught several supply chain attacks since I started monitoring, and am building an automatic response pipeline to validate and publish new attacks.
I’ve also been working on python support and hope to have that released soon.
RGBloom: create printed photo books in minutes (UK only for now) for just £10 including delivery. Been working on this for 8 months, recently launched iOS and Android apps to go alongside the web app. Got just a few dozen orders a week atm.
I'm focused on making it privacy first, keeping as little data as possible for as short a time as possible (no accounts, you don't even need to provide an email address, and if you do it's wiped in 30 days). And transparent pricing: not getting hit with large delivery fees at checkout, and no discounts just honest prices.
1. FormBeep(https://formbeep.com) at $38 MRR now - A form to Whatsapp tool. Stack is mainly CF workers with Clerk Auth.
2. SlashLink(https://slashlink.app) $0 - A simple pay as you go amazon link localizer. Great for affiliate folks and a cheaper alternative to Genius. Stack is same, CF workers with SQLite(D1)
3. StackStats(https://stackstats.app) $791 - A better advanced analytics app for Substack writers, fully local. I have been writing on Substack for a while now and their analytics suck!
I'm building an open source quantitative finance pricing platform built on top of QuantLib.
Think of it as a distributed QuantLib, exposing pricing, curve construction, risk analytics... through APIs, making QuantLib easy to integrate with applications without having to deal with the C++ library directly and an extra layer of abstraction on top of it.
A bit like iodine in the basic idea, but the execution and network topologies you build with it are different. It's more of a building block; you get your DNS channels represented as UDP endpoints, and take it from there, ie you can run Wireguard over it, or SSH or a SOCKS5 proxy (using quicmux, so over QUIC), or run Mosh straight over DNS if you're crafty.
It has an Android app too! I tried to submit it to F-Droid but they're so swamped in spammy or low-quality apps that they're not taking any brand new apps :-(
A sad state of affairs, that.
I have been working on building an agent and workflow builder that converts chats with claude, chatgpt, gemini or any other chat agent into a workflow or autonomous agent. Its live at toolscaled.com
Essentially I found the key thing with modern ai assisted development the key thing is to manage your own time/ parallelize without having to switch between chat session as much as possible.
In my free time I'm doing a lot of things, it's a little crazy—an Imperial Rome TTRPG, an archeological research RAG db to fact-check it, a browser-extension agent harness, experimentally training NCM models (eg entity extraction (didn't work), or playing Wesnoth (jury's still out)), US Cartograms (maps with proportionally adjusted areas), and Jug.fyi (a probabilistic distribution calculator).
Cloud save sync between a real Super Nintendo and emulators.
I’ve just got back into hobby electronics after a fifteen year diversion into software. I set up my workbench this week and this is my first project, something I first imagined building twenty years ago!
I spent the weekend wiring the spare controller port to a microcontroller to act as WiFi bridge. I have a flash cart, so I patch the reset vector to a small save sync routine that talks to the microcontroller.
It was especially magical having Claude read from a logic analyser and flash firmware tweaks iteratively. I still got to wield the soldering iron at least.
I've been doing the same, just with a llama-cpp local serve integrated + tor as a proxy. I was wondering whether there's a way to limit the heating issue with macbooks when running local model: do you think that limiting the number of cpu in the container is enough (albeit slowing down the llama-cpp inference speed) or is there a better way?
A small team and I are working on Sendery [1], a transactional email platform for product teams. We're trying to fix some of the headaches that come with managing product emails as companies scale: keeping track of hundreds of templates, maintaining consistent designs, coordinating changes between developers and marketing/ops, and managing everything across multiple languages is way more time consuming than it should be.
It's been a fun project to work on since it's been a pain point for myself on many projects, so I'm mostly just trying to solve my own set of problems and hoping other people find it useful too.
I’m working on a document library app for macOS with tagging, collections and user attributes. Unlike other apps that pull files into their own databases or reorganize files into their own folders, Librium indexes files in place, so your existing folder structure stays in tact. It works with folders on your local drive in addition to external and network drives and the cloud. Search works with on-device text extraction and OCR.
It’s currently in Test Flight beta, but I’m planning to have it to App Review this week.
Building Wishfox [1] for around 6 years. It's a wishlist app with price tracking, groups, secret Santa and more. Recently did a major rebranding, and the mobile app is almost ready for testers.
Then I'm working on improving my two open-source apps, Portfox [2] and Localfox [3]. Portfox shows ports for running services and dev servers, Localfox provides them with .localhost domains with TLS and Cloudflare tunnels.
I've been slowly working on migrating things to a more well documented setup as I build things. For the longest time I just set stuff up and then.... Hoped for the best. Started with my main authoritative DNS for some of my domains today, still working on some Ansible setups but I have the zones down in Terraform now. Is it overkill? Absolutely. Am I learning stuff? Also yup
The idea is that the source code is annotated with special rules which configure how parts of the code (also configs and docs) depend on each other. This can be done either directly in the source code comments or in a config file blockwatch.toml
This tool is very useful for large codebases that e.g. share constants across programming languages and configs, have docs that should strictly reflect the current state of the code, etc
It uses tree-sitter to parse the source code, 35+ languages are supported.
The project is actively developed, your feedback is welcomed! Thanks!
Unifying all the various image/video gen (comfyui for local, api for closed) in one cohesive desktop app, with integrated agents to pilot the whole thing (if you want). It's been fun to build and quite fun to use. Technically out in beta and testable, but working on refining onboarding/docs before I try going wider.
Finding a balance between deep enough to satisfy more technical creatives (like me) and less technical ones. Agent integration helps tremendously with the latter.
Still working away on my book (and library book) recommendations service https://leafle.org
It has if nothing else been very useful for me (it has correctly deduced that I would get something out of borrowing Grief is the Thing With Feathers, the Dispossessed, and Cosmicomics, none of which I would have thought of reading). Plus it has been a fun opportunity to play around with LLMs as an interface and a research and reverse engineering tool; as well as Jev for non-user facing quality work like banning crap openrouter providers based on their output before putting it in front of a user.
Most recently I’ve been dropping off flyers for it at a few local libraries in my area.
I think a lot of people have variations of this as personal software. I’ve just been tracking on requirements and updating mine. I don’t have grpc and JavaScript scripting but it does no code style assertions they claude does a good job of maintaining.
Let me know if anything glaring is lacking. I was wondering if people want the test suite results to a remote Postgres instead of the local SQLite but not sure it’s needed
1. Broadcast (https://sendbroadcast.net) - Self hosted email marketing software. Been working on this since 2024 and positive revenue. Releasing v3 this week! :)
APIKulture. An alternative to Postman/Insomnia/Bruno.
To be honest, I started this because Postman and Insomnia became forced-corporate-cloud not usable for me anymore, and Bruno lacked a feature I considered fundamental for my workflow.
The feature is to remember the JSONPath expressions between requests, a feature that Bruno has implemented already.
But I created this thing and I'm using it daily since then. C++ and wxWidgets, it uses 40MB of RAM, and it is super fast, instead of the 400MB used by the above mentioned programs.
I started working on a little news aggregator which has kinda spiralled into more of an open web thing. Keeping it open source, and open to feedback on how to better improve it.
I’ve tinkered with some decision model stuff too, and it can do things like identify possible spoilers in comments, or spiciness levels in comment sections. Works with HN, Bluesky, Lemmy, Mastodon, and can import RSS feeds too.
For reasons, my team is pretty much limited to Argo Workflows for our CI.
Nothing against Argo Workflows, but I really wanted to get out of the business of building our own CI product on top of it.
So, I’ve been working on this CI “framework” on top of Hera (the Argo Workflows Python SDK, which is really neat) that will hopefully make it easier for our team to build CI pipelines, with CI “primitives.”
I'm working on building browser-based learning games that help my 6 yo' to explore and understand small concepts through tiny games https://chalkolate.com
I build it based on what I thought daughter could improve and ideas that my daughter gave. I am currently working on introducing basic electronics.
As I'm working on my regular decision college applications, and since Opus 5.5 released, I was able to polish up Quillium (https://quillium.bryanhu.com) so I can dogfood it for my college applications.
It's a nonlinear writing editor, so I can keep track of all the versions of different words, paragraphs, sentences, or drafts. No more Final_FINAL_v3_essay.docx
Still plugging away on Flopper to build a holistic view of the GPU market including providers, rental prices, vendor information and mostly - specs for GPUs.
I've been building RCHIVE [0], a desktop app for cataloging, searching and reorganizing files on your computers and drives.
I've got 30 years worth of poorly-organized junk strewn about various computers, NAS, backups, folders, etc, and every year that passed made it harder and harder to rein it in to something sane. And last year I had one drive start to go out on me, so between all that, and consulting contracts drying up like a shrub in the Gobi, it felt like time to build the tool I always wished I had. Ended up rolling the dice and turning it into a real product too, took about 14 months (mostly full-time) to get from start to launch.
Anyways, it scans many different sources, gives you unified deep search across the entire catalog (drives online or offline), allows you to virtually organize and clean up giant archives of digital stuff, resiliently copies or moves files when you're ready to actually restructure on disk, or just lets you quickly pop in to find something / browse everything with previews, metadata, etc. It works on Windows, Mac or Linux, works offline, has zero telemetry, deletes nothing, and is fully deterministic in operation. So, like, no goofy AI integrations for decision making or classification, everything is predictable and repeatable, you can even run portable versions off of a USB drive if you don't want to install it.
I also have a companion app RCHIVE Examiner [1], which is a free ncdu-like tool to see where the hell all your hard drive space went, and what's hogging it. Currently only in macOS App Store, but Windows and Linux versions are releasing soon.
There are a handful of discrete, decoupled components underlying both apps as well that people might find useful (especially in CLI scenarios) that I'm also planning on releasing gratis as soon as I get a chance to package them up standalone.
I'm working on Zustri (https://zustri.com), booking and scheduling for therapists and salons. My girlfriend is a therapist and didn't like any of the booking/scheduling software she tried, so I built her one.
I finished the basics a long while ago. Since then I've added an AI agent, a public booking page, and support for multiple staff. Hopefully, I'm launching soon. Currently finishing final touches, gathering feedback and trying to figure out how to market this
A reimagination of bloodborne as a native VR game. All interactions have been redesigned to feel physical in VR, and combat retuned etc. It feels less like the original and more like a native VR experience now
My son is dyslexic. So I built a Chrome text-to-speech and reading tools extension to make it easier for him to get access to written content online (and also to support him at school). It has support for Google Docs and PDFs as well as the web. There are a few of these, but this one tries to stay out of the way, have better tracking, better voices and lower cost. Still more to do on the reading side - writing is next after that.
Serverless Wordpress + MariaDB without requiring plugins or any modifications to customer code: https://www.agiler.io
The serverless MariaDB is important because it can scale to zero, so you have dev sites that cost less than $0.50/mo.
Also added MCP and CLI tooling that exposes all of the infrastructure + WP CLI + DB so agents can develop WP sites... which is nice because you can still use the WP admin interface, while using AI to develop the site, debug it, change pages and posts, etc.
Magic AI - I'm using it as a sanitized kids chatgpt for 7 activities: Story, image creation, chat, reflections, journal, learn and quizzes. Sends me daily summary on Whatsapp.
It started with building pi in the cloud - with pure cloud flare workers. I have it working now, and can spin up - the wrangler agent can fully run it as long as you pass it an inference key, including a file view, a chatgpt like experience, and ability to build and host custom apps. Can use alarms to wake up to do plain language tasks, etc.
Trying to figure out how to package this up, and where this would even be useful for other apps.
I needed a code-first approach for making installations and interactive art. MaxMsp gets messy with very big projects. Python didn't really work well for music making and Tidal/Pi/Sclang are only meant for sound. So I needed something that has bpm synchronized threads, but can draw and do anything else like file IO, networking, hardware control etc. It works quite well now.
I keep saving blogs and articles but almost never revisit them.
So I’m building an Android app designed to help me actually read what I save. You only get one saved article per day. Finish today’s read to unlock the next one. You can also save Twitter, Reddit, etc. posts separately and revisit them anytime.
Right now I’m mainly validating whether people would actually want something like this. If there’s enough interest, I’ll run a small closed beta for Android.
Hunting bugs in the Python ecosystem using fuzzers and AI tools I develop. If you have a Python package or extension you'd like analyzed, just let me know.
I'm working on a better way to design longer scrolling pages with multiple sections, while preserving the visual look/feel and drastically speeding up the experience.
As a disclaimer I cut 32 seconds of generation time out of that video, but still I'm quite happy with the speed. Not launched yet, but it's been fun to work on / wanted to show off progress :)
Writing a Newspeak implementation of the PostgreSQL TCP protocol!. Talking to a GnuCash database from Chrome web browser. Uses my Newspeak TCP library. Both are bare minimum, probably very broken, but I am able to parse a Postgres result set from a simple query.
Here is a one minute video where I print the result data to the console from a Newspeak workspace.
I'm working on https://github.com/BowlingX/super-release, an alternative written in rust for semantic-release and mono-repo support. I build it mainly to make shipping packages in larger JavaScript/Typescript mono-repos easier. Also I started getting into rust more and wanted to learn the language.
Still changing along on Incredible (https://www.incredible.rs/) a Rust Text User Interface (TUI) Framework for the 2nd Quarter of the 21st Century.
It is New, Experimental, Modular, Declarative, Event Driven and Reactive. It builds cross-platform apps that can be compiled for the terminal (macOs, Windows, Linux), a standalone native window (macOS & Windows 11), or the browser (via WASM) from one unified code base.
Curious if this project will ever be 90% done.. ;)
HostsTray, tray utility to manage your hosts file, that also has a reverse HTTPS proxy builtin, that installs a self-signed CA and generates certs on-the-fly.
Also it forces your browser to "flush" their DNS by killing the keep-alive connections that for example Chrome keeps open in background. And it can be controlled by your local coding agent through a simple API (which also serves the skill for it)
I’m building Tinta, a journal app for Android where all the AI runs on the phone. It works in airplane mode once the models are downloaded.
You write or dictate an entry, and a small model (LFM2.5 1.2B) asks you a question to reflect further. It turns out 1.2B models are decent at asking questions, so I leaned into that. An embedding model finds older entries about the same thing and shows bits of them under the new one.
Most people using it are on budget phones running the models on the CPU, where the app provides good experience.
Great idea! I'd love this, but as an open source app. Out of interest (not a criticism), what made you decide to do this closed source, especially considering there doesn't seem to be a commercial interest involved (for now)?
Thanks for a good question!
I see this app as an MVP that needs user feedback to deliver exceptional value in a world where cloud AI is abundant and doesn't use 2GB+ of RAM to run. I am excited to see what shape it will take, and it feels too early to determine source availability.
ColliePWA[1] a mobile interface to continue working with your agents when you're on the. Trying to meet devs where they work on the desktop in multiplexers (herdr/tmux/zellij). Remote access can be set up with any front door like tailscale.
Still not sure where this is going as the space is super crowded and moving fast, but it's solving a lot of my own problems and doesn't rely on third parties to run.
Privacy friendly open source browser extension to export chats from web client of Microsoft Teams. Reached ~90000 users now. Currently working on making it more stable to export huge chats, which contains several years of data, while keeping things efficient and fast.
For the first time in along while, I'm not actually working on anything. I do some vibe coding for fun.
I'm kinda leaning towards vibe coding my own GUI coding agent. I currently use Pi and recently have been playing with Herdr and Paseo, but none of these feels right to me. These are amazing tools, but ai think we could do better. But I'm very hesitant - I know the only way to "succeed" in a project like this is to write (vibe code) hundreds of thousands of lines of code and it's kinda depressing.
Getting https://streamish.github.io/ ready to launch, a music server compatible with multiple 3rd party mobile apps and particularly Synology and QNAP. Want to do Plex next but I suspect that's a dead-end since I think you can't directly sign in you have to do it through their website.
Now putting it to personal use and keeping an eye out for issues and rough bits to smooth over.
I've built a habit of completing a game of sudoku every night before bed. After being frustrated with how messy my board can get, I decided to build an app for my reMarkable Paper Pro. In my opinion, it's a big step up from both paper and phone apps.
Local-first archive for AI conversations for providers without sane exports like Google AI Studio, OpenRouter (including chats), LM Studio, and others, as I couldn't find any chat archive with OpenRouter/Google AI Studio chat support. It supports semantic search, MCP integration, and continuing old conversations with llama.cpp.
npx threadshelf
https://github.com/ChrystianSchutz/ThreadShelf
Recovering from pushing out an article on coroutines [1]; coordinating getting the house in order for the workweek [2]; and drafting a package to make literate programming in org-mode a bit more to my liking.
Over the last few months, I decided to learn more math and dive deeper into LLMs, RL, and robotics.
I ended up building a personal agent that asks me questions every day, emails me interesting reads in these areas, and helps me progress in my learning.
It started as a web app, but I have just ported it to a Mac desktop app.
It works like Conductor, but for your learning side quests: https://miletus.app
Rebuilding my business’ website with Claude. It’s gonna be 10x better and migrate off Wordpress. My web designer broke something and stopped answering so giving it a shot and it’s going great so far. So stoked for the final vision I have.
I continue to update my coach app that generates workouts for me. It’s really solid at this point, better then the trainer I had, so mostly occasional bug fixes now or silly additions.
A new kind of grid-based flowchart editor. The design is to have a grid instead of a canvas, and all nodes are the same size and sit in a cell. On top of that, there’s keyboard navigation kind of like a spreadsheet.
The idea is to make editing flowcharts way faster by being keyboard centric and having autolayout.
The project lets you find out which agent might perform better in your codebase. It scores the codebase for different agents using their static signals and gives you a response.
I'm putting the finishing touches on a new website for my LisaGUI project. The design of the new site is much more modern than the current site's design... I just have to finish rewriting the copy on one of the pages, and there are a couple of minor things to add... and then it will be done!
A better, model-agnostic artifact experience to work with AI visually. You can annotate and draw on artifacts, so can your agents. You can share the artifacts with colleagues or friends and with your permission they can send comments to your agents too.
When you need more compute, artifacts turn themselves into microVMs automatically (paid feature) with a 8GB allowance
Currently we are working on Kumoss Open Source, intent-based infrastructure platform that turns natural-language requests into infrastructure changes, while keeping validation, compliance and existing engineering workflows in the loop.
Basically allow developers to request any infrastructure need in a web portal and automatically that request is fulfilled following the organization standards the platform team defined.
At its most basic it keeps track of all your SRS-y things for you and builds stories in worlds you can create to help review them, but it's so much more. It essentially combines a bunch of things I've wanted as a language learner with my favorite aspects of various apps I use.
Signal analysis to drive <desired signal-driven action>. Grouping/categorizing/stratifying signals into useful buckets to drive actions. Blind source separation has drawbacks like continuously guessing at sources and labeling them as brand new instead of having a memory of previously identified sources. Think of how humans can distinguish pianos from drums and remember the distinction across songs.
I have a service that crunches public filings in the financial sector: https://embarc.com
There is a lot of data for browsing, though it doesn’t aim to solve specific user problems as of yet (I did take a stab with RTOs in the logged in dashboard).
Artists make users jump through hoops to like, comment, repost etc.. before giving them access to the download - CrateDiggr highlights the free downloads in the Soundcloud UI and clicks through the download gates for users.
Been a fun passion project, check it out if you download tracks from Soundcloud!
I've always been interested in health, longevity, and biohacking.
I'm Canadian but I moved to USA to work for Microsoft. One thing I liked about the USA was how easy it was to order bloodwork/lab testing. My original idea was to bring that experience to Canada, but since starting I've found other ways to improve the patient healthcare experience in Canada too.
Still cracking on with https://whistle-enterprise.com and making improvements to the document generation model. I'm ditching my old trusty Phi4 model and testing out some other small models to fine tune, something about Phi4 just seemed to work quite well for what it was so it's a hard contender to beat for midrange laptop CPU inference at a decent rate and quality.
Its like coolify but cooler since you don't need any technical knowledge to host apps. You can also run claude and codex remotely on the vps without keeping your computer or laptop open all the time.
I'm improving https://davegooden.com & just pushed https://edinacountryclub.com live as I continue to build out my real estate empire. (kidding, not kidding) Hopefully the industry can hang on for 10 more years.
https://www.haptrix.com - An Core Haptics editor, with the release of the iPhone Duo, the editor finally makes sense on device, so I've been working on the new UI.
https://www.phonique.com - An AVAudioEngine/PHASE Audio graph editor, it's easy to connect nodes together visually, then export the wired graph as Swift
I built Rindraw, a (hopefully better) Procreate alternative that runs on mac, ipad, iphone.
I believe that everyone is and will become an artist in the future, and everyone should be able to express their own unique ideas thru, especially in the age of AI
We just got a puppy, and needed a way to track potties to help us with house training. So, with AI/Claude's help, I found a similar open source package (Baby Buddy), forked it, and very quickly made it my own. Self-hosted, local for privacy reasons.
I'm working on a TradingView Pine Script compatible Python framework: https://pynecore.org . And a Pine Script compiler for it + a full IDE for Pine Script with syntax highlight, language server, full featured debugger, charting, data downloads: https://pynesys.io .
It is 99.771% bit exact on the top 1000 Pine Scripts and 100% trade match.
I'm doing it solo.
I am building a tandem, reverse-trike halftrack, using the front end of an ATV and the back end of a snowmobile.
Replacing snowmobile skis with wheels is not a novel idea, but I wanted something which rides more like a motorcycle, with extra length for a passenger and a cargo basket.
It should end up looking like it came out of Mad Max; I plan to cruise around at festivals, running an ad-hoc taxi service, as a way of meeting people and striking up conversations.
* files in organised trees/graphs generated using LSPs and symbol references instead of linear file list
* keyboard shortcuts for everything from navigation, pan, zoom, comment, note, submit...
* mostly github compatible with native features
* LSP powered symbol resolution for inspecting definitions and references
* LLM powered reviews supported
* Inline LLM question answering
I've been building out a link blog idea called Linked List that I've wanted for a long time. Curation will be even more important in the future and I would love to have another option beyond Hacker News and Lobsters to find good reads suggested by fellow humans. It has been a fun space to explore ideas.
Not quite ready to open up sign-ups, but you can get an idea of the site here: https://linkedl.ist/@kale
I've started maintaining my own Firefox fork that:
- removes the stuff that I don't care about (Firefox sync, VPN, etc.)
- fixes a lot of the awful UX/UI choices that the (frankly) incompetent product team at Mozilla refuse to
- actually adds stuff that I do like integrated mail and calendar, a document library system (basically Notion), native ad-blocking, terminal, task-tracking like Trello, time-tracking like Harvest, etc.
- has integrated AI (Apple intelligence for MacOS for the easy stuff and Claude for the hard stuff
It means that instead of having half a dozen applications open, I can just work within my Firefox fork. Because it's all deeply integrated, I can do wild stuff like:
- tell my AI to search the web on a schedule and send me a custom daily brief with my own email
- maintain a comprehensive memory of all of the work that I do, across half a dozen clients, and be able to query it easily
- pull analytics from multiple web dashboards and format them into something cohesive
- program the web with a node-based infinite canvas that has live websites as part of the workflow
- manage my tasks across multiple domains (for example, it will see that I sent an email to a client and automatically update a task that says that I need to do that)
It's incredibly powerful and it has made me much, much more efficient.
I'm creating an IDE for founders to run their operations. Create automations (deterministic), dashboards, and soon notification systems. It's a desktop app, and the whole IDE runs in pair with claude/codex through local mcp.
It was developed as an internal tool at my other startup, Ipek helped us to run the whole thing smoother and with all the context I need.
been curious about the intersection between observability, SDKs, and product analytics. i couldn't find anything around SDK observability so built this little tool. curious what yall think!
i assume at bigger orgs there is some internal handrolled tool like this that ingests datadog/sentry/whatever info...if anyone plays around in this space and has feedback id love it!
I am building a system that connects texts (books, research papers, etc.) in a meaningful way to help people decide what to read and be aware of how texts relate to each other.
I am building Abbox trying to help engineers focus longer on tasks without being interrupted by non-technical people with small requests.
I always got annoyed by change this h1 to h2, add tracking to share button, why can’t the customer play the video, the styles don’t match figma.
If you are engineer who also gets distracted I would love to hear your opinion and feedback. It is a bit hard to put it all together as every department has his own needs and opinion about the website.
I'm still working on https://openaltfinder.com, a directory of open source alternatives for proprietary software.
Trying to find out why Google has completely de-indexed the website, but other than "low quality content", I'm not getting any answers. Even Bing brings in more traffic...
Writing. I'm going to start an MFA next fall at LSU New Orleans. My latest piece, Three Black and Milds, a night in Juarez. No commentary, no lesson, just the city and the people in it, what it sells and what is bought.
I built the original for myself because I wanted to listen to acoup and other history blogs while sleeping. And in the spirit of getting better at finishing projects decided make it public even tho it’s nothing really unique or special.
Still lots to polish especially in the text/copy and ui but the core features should be working.
I'm building an attribution and analytics tool for onchain apps.
It's been a learning journey into data engineering and business intelligence best practices. Thinking of taking my learnings into a new product for a different segment!
I've been working on an Obsidian plugin to quickly enter caloric and nutritional data (https://www.github.com/Krishna-Sivakumar/joule). The core features are done (A DSL and a runtime for finding and resolving formulae recursively), and I need to work on documentation and some convenience features. Hoping to finish it this month!
It uses Apple Intelligence to help brainstorm domain name ideas. I’m interested in seeing what small local models can do as they catch up to larger ones, this is my first iOS app so I started with something simple that scratches my own itch.
Maintaining my indie social network for tech people https://stackness.dev/ (currently struggling to keep users on the platform, even games and achievements didn't help, can you believe that? :D)
and this weekend I'll finally launch a tool for guitarists and their friends https://kantamigo.com/
I mostly focus on solving my own problems, and if I find a solution that can serve other people, I sometimes push it out as a product.
1. A watch app for runners that focuses on competing against previous runs, inspired by the rally games of the 90s where you could race your "ghosts"
2. A LEGO deal tracker that focuses on collectors and LEGO investors and is focused on price prediction based on data.
3. A mobile tracker for all the hundreds of components I use when tinkering with physical devices.
...
(Disclaimer: all of them are built using AI-assisted coding, and I wrote almost no code at all, although I have 30+ years of experience as a developer)
Working on a natural language intelligence layer for contacts - "Pull everyone I know who works at a Series A healthtech in the Bay Area who also went to Stanford for ungrad" => either structured or unstructured results. Large-scale data support, automatically enriched with up-to-date data, fast/accurate results. Integrates natively with Gmail and Gcal.
Looking to pilot for free with small qualified teams (early-stage startups with real/immediate need).
I'm reverse engineering the streaming protocol for three different brands of cheap surveillance video cameras. Each brand has its own app. They're all awful. None of them have a native desktop MacOS app. I've successfully created a native app for MacOS and Android for one of the brands. They're faster and work better than the ones they're replacing. One down, two to go.
A suite of own cloud based local first apps optimized for eink tablets, specifically the boox palma 2. Notes, book reader, browser, music player, maps, camera etc. With a companion ios app combining all of them.
this way when i want to disconnect i can just grab the palma and go
When I was a kid I loved a game called "Stars!" from 1995. I cloned the core and I'm also adding my own touch and lots of high sci-fi concepts. Already fun to play, early access coming soon: https://store.steampowered.com/app/5482440/Asteria_4X
I've been working on a voxel game engine called Bonsai for ~10 years.
Probably the most interesting thing about it at the moment is the editor. The world, and most things in it, are represented as collections of SDFs. More accurately, they're density fields, but, potato-tomato. One neat thing about representing the world as SDFs is they've been thoroughly researched and documented by a guy named Inigo Quilez, and they have a lot of nice mathematical properties. For example, you can do a smooth union of arbitrary SDFs to get nice rounded contours where shapes join.
The world generator and editor operate almost entirely on the GPU. The terrain and editor pipeline is a three stage process. The first stage is user-configurable; users write an arbitrary 'shaping' shader in GLSL. The second stage is a fused stage; it does a convolution on the output of the first stage to compute a normal, runs a second user-configurable 'decoration' function, and finally applies edits. Edits are collections of SDFs that are 'compiled' from a graph structure on the CPU into a bytecode, and interpreted in the shader. The last stage runs another convolution to compute the final normal, then packs an occupancy bit, color, and normal down into a 32-bit representation, suitable for transfer back to the CPU. I expect I will expand this to include additional user-level bits for encoding biome type and GPU-driven spawning.
The entire shader pipeline is one fused 'ubershader' .. everything is bound at startup and the shader is dispatched three times, incrementing a 'Stage' uniform on each dispatch. This strategy conceptually makes the driver overhead nearly zero, and the profile agrees.
The final step in the process is giving the CPU a chance to spawn additional entities and structures after generation. This is not completely fleshed out, but the bones are there. The idea is that the results of the GPU work are passed back to the CPU, which can run arbitrary callbacks to spawn additional edits and entities. Once spawning is complete, the result is meshed using a fast binary meshing algorithm and uploaded to the GPU. I keep 16 bits of data per voxel from the GPU work, 1 bit of occupancy, and 15 bits of normal for collision response, which works out to half a megabyte per chunk. At the moment I only keep chunks which contain a piece of the world surface, and discard the rest. Surface chunks are dense, and may be compressed in the future, but keeping this amount of data scales to world sizes greater than 1B voxels cubed, which seems good enough for now.
In terms of raw performance, GPUs are ridiculous. My 3080ti mobile gets somewhere on the order of 500B-1T perlin samples per second with a naive perlin implementation. Compared to FastNoise2, that's about 1000x more, which sounds wrong since my GPU only has 35x more capacity, but, that's what my initial profiling said. That number is the marketing number; the real throughput is significantly less due to multi-stage nature of the system, but, there's a lot of headroom for improvement. For example, dropping the backing textures from 32bit to 16bit roughly quadruples performance. Since we only really care about the surface where values are near zero, I believe using 8 or 16 bit is a viable option, though this claim will require a fair amount of study to validate.
All in all, I'm pleased with how this is all working out. My original goal was to write a highly flexible system giving users both easy-to-use tools and granular control over the final output. I think I've done a fairly good job at that, and the fact it's also super fast is the cherry on top.
I'll do some shameless self-promotion and leave some links here for anyone interested in looking at the engine, language code, or some pretty pictures.
That's so cool, building something consistently for 10 years is no small feat. How did AI helped you, or How did AI impacted your day to day work in recent years ?
Thank you :) One of the nice things about working on a game engine is it's actually a large collection of small projects disguised a big one. That let's my squirrel-brain jump from thing to thing and have enough variety to not get bored.
I just started using AI a few months ago. My smoke-test was "can it write a simple memory allocator". Until recently, every time I tried to get an AI to write an arena or simple buddy-style heap allocator, they just vomited all over themselves. To my surprise, the last time I tried, the AI absolutely crushed it; it unearthed tricks I didn't even know about in one shot. Since then, I've been using AIs to do the boring, rote work that effectively just boils down to me typing in something that's in my head. For that, they're great. I typically sketch out the thing I want it to make, then tell it to fill in the blanks. Sometimes they still go off the rails, but all in all it's a useful tool.
Where I've found they still struggle is making any architectural decisions, or where performance matters.. even a little bit. They often make super braindead decisions when it comes to performance for no discernable reason. They'll course correct when you tell them to, but I've had to keep a close eye on the output and reign it in frequently. Asking an AI to design a system from the API level hasn't ever returned a good result for me.. I always do that work.
All in all, I feel like using AI will probably make me a better programmer in the long run, contrary to popular belief. They accelerate the boring stuff I already know how to do, which leaves me with more time to work on the edge of my skill level where I grow as an engineer. Guess we'll see in 5 years.
A video game in which players can build their own worlds. The brains of the creatures in the video game are a small language model + harness that runs in your browser: http://id-le.com
I'm still working on my own shell based AI tool. This month I am testing and reworking it to try and get useful results with really small models, like 2B level. Goal is to get help debugging Linux issues when offline with these small models that can run anywhere. Then also being able to access cloud based models to do harder stuff once online. all while feeling like a CLI tool.
The work mainly stems from a belief that Nvidia gpus' architecture are a bit subpar for machine learning and that tpu could be a viable alternative but are still rather undocumented.
I've been building data pipelines from sometime as part of some other products, eventually pulled out a part of that system into a standalone product called hydrafetch.com
That's the main thing that I'm working on these days
Working on a tool to provide a thin OIDC auth layer for self-hosted LLM engines. Users log in with your Org's identity provider, generate an API key for themselves, point their harnesses at your instance of the relay, and then they're off to the races.
No web chat interface, no call-us-for-pricing feature gating, just a tight Erlang program for people who want to get shit done!
I’ve been working on my inertial navigation library. It actually beats ArduPilot’s navigation subsystem in my benchmarks. I’ve been working on these algorithms since my PhD thesis and now I am releasing the stuff as open source embedded C and Python library:
I'm building mensara, multi player memory so you're whole team can work together with ai coding software.
One of the most fun things is you can use all the various agents, like muse, grokbot, dots etc and they all know you immediately and can work together.
I've been working on a small search engine for content creators that play and/or review indie games. It's written in C and has been a lot of fun to work on.
Infrastructure for hosting static webpages with options for storing data, having users, sending notifications, shared "rooms", and various other bits and bobs all behind an mCP. All $0 practically because it's made up of static services from AWS.
Made for me, but I can provide access to friends and family, which is nice.
My integrated terminal + file browser with enhanced autocomplete and search. Replaces all native OS ones for me: https://github.com/David-OConnor/peregrine . Also includes misc utility OS functionality, tailored. (Advanced clipboard, a scratchpad etc)
I started a company to focus on the niche accounting professional service called Cost Segregation in the summer of 2024. I was laid off from Meta in February 2025 and dove in full-time on that startup, Segtax.
We are on the cusp of automating the bulk of the work. It's been an exciting ride
I've just released v1.0 of Rouler which is VPN split tunneling for windows. Think IP rules for Linux, but with per-app/service connection steering. The whole app is built in Rust, including the kernel driver. Having released v1 I'm now starting work on adding folder and windows store support. https://rouler.io
Polign Recall, a state and memory management for agents that can provide append only state logging and governance. The memory and storage is completely on BYOC setup and can work with confidential workloads
I’ve been working on an accounting agent factory (https://www.translucent.io) .
Building a harness and infra that can process books with thousands of invoices, transactions, etc. is quite rewarding and challenging. In particular the long-running nature of things.
Still working on ShopSpec, a parametric deterministic furniture builder. Trying to find time to put more actual builds in the gallery - have a few projects that are in progress while continuing to build out the tool and iterate where I find personally useful.
I have spent the last six months or so on a multiplayer, open world game. The idea is you can jump to any city, walk around, see businesses and chat with others. Loading the resources (map etc.) as they are needed and garbage collect have been a challenge,
Building a soccer game based on Mike Tyson's Punch Out, just to see if I can create some method to see if an agent can iterate on making something more fun. Obviously that depends on the definition of fun (which is already a can of worms), but it's a fun way to use a different side of the brain.
The machinations of my own mind/psyche/nervous system. For various reasons, I have had to get increasingly curious and stay that way over a years-long haul. This includes the usual suspects:
- "Serious Stuff", lets say. Viz. paying attention to and learning about therapeutic introspection, relationship/social dynamics, clinical psychology, meds as they apply to me.
- BUT also can include "fun stuff"! For example: seeking out and befriending like-oriented kindred spirits (it feels great to be seen / heard by peers), and developing a persistent interest in biology (which I hated in school), and even signing up for research studies! I got accepted into a Phd memory study and it was great to be back on a university campus, a couple of decades since graduating.
Current opinion: I think all mental health work leans N(1).
The art of it appears to be remaining curious and seeking self-understanding without lapsing into tiktok pop-psych, nor a false sense of studied erudition (No; one will almost certainly not become a clinician).
The greater the complexity/depth/intensity of one's condition, the more it calls for subtle, bespoke work. Progress from outside-self intervention has an upper limit (e.g. "taking the prescribed meds"). Active, self-directed self-work (in concert with one's support system, of course), brutal as it, must be done... somehow.
To others in a similar state space; may The Force be with you. Following one's curiosity, even now, will probably not result in the curious cat's fate. Worth the extra (omg so much extra) trouble, IMHO.
_|/
P.S. Anyone wishing to discuss more, feel free to email me (see bio). Please refer to this exact comment.
I've deliberately not shared specifics, including self-educational resources, because this stuff is slow, plodding, deliberate N(1) work, and I've personally found it counterproductive to read/consume the vast majority of N(1) experiences I've found.
For the month of November (and starting in October this year) I like to hold weekly cooking challenges to practice and refine my Thanksgiving day feast. Now about 3 weeks out from the final boss level of cooking, and I'm having a week all about brussel sprouts. Earlier weeks were cauliflower and the classic seasoning blend of sage, rosemary and thyme.
A few projects, but now waiting for a week for my Piano Game[0] to be approved on Play Store. Maybe it takes longer cause I marked it as being "for kids" too.
An Obsidian plugin that turns handwritten reMarkable notes into searchable text. You can run the handwriting recognition locally on a Mac. The latest feature I added sends your highlights and margin notes from the tablet to Readwise. https://taggedsync.com
Building a native macos app with a suite of completely local AI tools. The first version is out and have a long roadmap in front of me. Would love to hear some feedback.
Building the web-embeddable legal IDE for transactional AI-native law firms. Docx, PDF and Markdown in the browser with redlining, go-to definition and more. In Rust, of course.
It combines nonograms with crosswords, requiring both logic and word knowledge to solve. I am still working on the interface and tutorial, as well as the puzzle difficulty levels.
The TLDR is that if you like Honeycomb or Datadog but don't want to pay insane prices, you should check it out.
The longer story is that I love talking to SREs, platform engineers, etc about observability, so if you want to kvetch about it, please send me an email at jack@telemetrymachine.com!
My cofounder and I have been working on observability for a combined 15 years and we decided to take what we love most about all these tools. We've got the powerful querying of Honeycomb, the great UI of Datadog (though most users use us via agent anyway), and the speed of Clickhouse.
Basically, I hooked up an ESP32 camera to a vision model so I can prompt it to watch stuff and send webhooks. Thinking analog gauges, legacy equipment, sourdough starters (still no API), and stuff like that.
Been dogfooding the prototype on my desk, and now I have an air quality API that reads the display on a cheap monitor.
Also added direct camera access via MCP so Claude could help me debug the hardware, which has been pretty fun.
cool stuff. I have cameras watching some animals and I've been working on real-time analytics for their behavior.
It looks like this is sending video signal off into the cloud? Are you sure people will be OK with this? Can I integrate it with local inference and lock it down?
Why can't this just run on a spare cellphone? Why the dedicated hardware?
I'd be curious to see some technical details on how you're routing b/w the user's prompt to the continuous monitoring. Are you purpose-building a tiny CV model for each prompt? Or are you running a big model continuously.
I took on a software project for a factory. During work hours, I write code for the company, and after hours, I work on code for other companies. I also submitted some PRs to open-source projects this month.
Working on a modern implementation of a torrent client [1], it's already pretty usable for most scenarios. Provides both a daemon + web UI out of the box as well as an embeddable engine for any Rust application.
Building my an airflow pipeline focused on collecting news.
Partly because I want one, but also because it makes for a good learning platform. e.g. initially do some of the processing with a stronger llm...save result and then attempt to train a SLM or similar
I am building Cloud based context compiler and intelligent codebase visualizer for Developers and AI Agents, for easy PR review, onboarding and easy architectural understanding
You can connect Claude desktop to databases using MCP servers, but Claude desktop is not great at displaying large volumes of data. A custom client can do a much better job.
It is a full-featured reporting tool, and can be used without AI. If you do use AI it has a semantic layer to help AI understand your table and column names, and relationships between tables.
I'm vibing an old style RTS game and exploring how far can I get with local models for drawing, 3D models, rigging and sprites. I love how much I can accomplish with Claude in terms of code, but gluing it all together with graphics still requires plenty of "manual" work.
Werkbook (https://werkbook.app), digital forms and photo reports for field crews. I'm a solo Rails dev, and thanks to Claude it now has native iOS and Android apps instead of Hotwire Native web-view wrappers.
Ur is an (very in-progress) universal formatter and linter for every programming language. It's in preview right now, which just means friends are messing around with it and having a good time pointing out all the bugs. Docs are mostly up to date, but buyer beware. Implementation wise, it's the Rascal meta language reimagined in Rust with some features from JastAdd thrown in there for good measure. Ur is incremental/persistent between runs, it's built on top of Mambo, a library that drops in as a replacement for Salsa and persists all queries to disk automatically.
Ur has some pretty good static analysis and code rewriting tools hidden inside of it that'll get documented later on. These tools have been developed in tandem with a static mass rewrite of some core parts of the Rust compiler. It's very much not done, but it's made pretty good progress towards putting all crates in one process, dropping memory after a crate is done compiling, and full incremental support across all commands. Here is a literate program that explains why, how and which changes are being made (I haven't read all of it myself yet, lots still changing). Aiming to have this out by end of the month.
Been working on my own Orca-like meta harness: getlattis.ai
Use it every day, and my team mates have started using it and loving it. In my dreams it’s a business. In reality the feedback I get from people that like it absolutely makes my day.
https://geotastic.us/ An AI analyst for ArcGIS. A lot of city data in the US is stored in ArcGIS or Socrata. This tool uses LLMs to generate code to analyze it.
I’m based in Berlin and currently working on ChatFrom.io . You upload your knowledge base, and it uses advanced RAG with hybrid search and reranking to find answers when someone asks a question. It connects to Stripe, Shopify, and other tools, and works on Telegram, Slack, and more. Free to start.
For the last year I’ve been building personalized tools to help my wife with her day trading and have started to productize them as https://cuppa.markets/
A desktop app to easily browse, tag, and rate your images. It doesn't touch your files, it just reads them. Everything is stored in a local SQLite database. Think Eagle app but private, offline, and easy to use.
https://hatamoto.dev - a complete replacement to Wordpress and its most used plugins built into the system. One go binary, no node , no compilation, Postgres and sandboxes plugins. Built in AI analytics
I'm working on a platform for selling event tickets (kind of an alternative to Ticketmaster or weezevent).
I'm working on it on my free time for now (I am also using it to test working with AI tools ).
Not sure yet if I want to build a company or not.
It's a free job site with some resume tools bolted on. Lots of people from all over the world are using it and that feels great.
It has 2.7M+ jobs and automatically removes any job older than 7 days. We never allow reposts because, frankly, the job market is a nightmare and any job that has been up for 7+ days almost certainly has 1000+ applicants already. You're just feeding your resume into a wood chipper.
Aficionado (https://aficiona.do) - a place to discover, scrapbook and remember media (movies, video games, tv shows, music and books) that resonated with you or that you'd like to try.
and on the side, trying to develop ai interfaces that aren't built around conversations, currently in the form of a restaurant search tool but i can't get it fast enough (at a low cost) to feel usable
working on AudioWeave, an ad free, offline first Android podcast player.
switching from iOS, I wasn't happy with the existing podcast apps, so I built this mostly for myself. I've been using it for the last six months and finally pushed a public beta.
My day to day is BE so I used the project to learn Kotlin and Jetpack Compose. Everything is strictly native using Room and Media3/ExoPlayer. I wanted a nice looking (subjective), privacy-focused app without any tracking or ads.
fxdax - https://github.com/semistrict/fxdax a port of fuschia fxfs to linux on top of a novel virtio device that stores file
data in object storage. Intended use is file system for VMs that supports dax to avoid page cache duplication while
maintaining high performance
Building out an interactive fiction platform which blends AI storytelling, images, video, and TTS with user choices at the end of each page in the story. Https://cyberstory.ai/
It started maybe 9 months ago with the idea of an AI driven Oregon Trail adventure and i've carried on with more historical fiction adventures across time plus alternate histories, romance, and more. Users can also build out their own stories.
One strange thing I ran into was the LLMs always naming characters the same names across stories (and i'm not using seeds!).
I've been doing a bunch of differential analysis and trying to put some of my large amount of private unpolished work online. Found a bunch of bugs in PNG libraries, Game Gear emulators, SDCC, and other things. I've been running a bunch of test ROMs on multiple revisions of hardware, and oh boy is that thing poorly-understood, if emulator errors are anything to go by. SDCC is the compiler I'm using for the ROMs and it's full of miscompiles and performance issues, so I'm running a patched build with everything I know about fixed, because I kept running into stupid issues while making the aforementioned ROMS (downloaded the whooole issue tracker and even fuzzed it a bit). Also running a patched build of Mesa after running into several bugs when trying out mesh shaders, a patched Fossil (https://fossil-scm.org/) because of bugs I randomly hit, a custom printer driver because the stock one sucks, custom firmware on my Siser Juliet cutter/plotter machine, ... and of course, barely any of this is anywhere online, and at least half of it isn't anything I wanted to do in the first place.
Anyway, I haven't shared this anywhere because I don't like how much AI-written wording it has, and I haven't really done any of the self-written analysis I wanted to put in there, but you can see the PNG analysis results at https://wizardmaxx.ing/difftest-png/. It compares against the 3rd edition, and you can select the 2nd edition if you want to for some reason. It's still missing an option to show what you'd care about by real-world judgement, but ¯\_(ツ)_/¯
I was recently laid off, so I've been pouring myself into a new paas offering and communications suite. Hoping to have something to show off here very soon. We will see.
i've been playing with this silly "game" inspired by one million checkboxes except the goal is to sort the board. something to kill time https://human-sort.app i like to think it's the same sort of work that's being done on severance
My main project is trying not to lose my mind, given all the things I can now create thanks to AI.
Jokes aside (although I'm not entirely joking about my mental health), my main project is https://auradistill.com: First-principles thinking for your AI agents.
It's not just about giving your personal AI agents "memory". It's about encoding memories in a way that can handle ambiguity and contradictions.
That project alone has got me completely hooked on building for and with LLMs.
I've got other projects too, but this is probably the one I've poured the most time and energy into.
A statistical analysis suite that runs in the browser via webR. It provides step-by-step guides for running tests and insights on results containing statistical pitfalls. Aimed at students who struggle with traditional stats apps.
working on talk deep https://talkdeep.me, a little app for keeping in touch with friends after i moved countries. you pair with a friend and get one question a day, or go through decks of deeper questions together. i've never done an app before so building something from scratch has been a really fun creative outlet
been building BuildOpsy mostly writing about system design, infrastructure failures and the occasional “well, that was an expensive outage” postmortem.
Also building calculators for things engineers usually calculate in a spreadsheet at 2am. you can see some on the site(https://buildopsy.com/calculators.html)
Not really a "thing" but I'm trying to come up with effective strategies for LLMs to generate screenplays that are not terrible. I think I have some that seem to work.
There is already a site here where you say what your job is and garages submit their offers. They can also auto-submit offers if it’s something routine like a standard service.
I feel like this is the third one of the November thread!
That said I will say again, working on http://leafle.org my app (mostly for me) to find new books to read that I wouldn’t have thought of (mostly recently Cosmicomics) based on a deeper understanding of what I like in books than people who read X also read Y, that I can actually borrow right away from my local library. Works on ~6,500 public libraries around the world.
Fun project to play with reverse engineering using LLMs, and jev and things like that.
I got tired of having newsletters in my inbox that I never read, and emails building up where you have to hunt for the important part, and not really being able to get the gist of whether the email affects me or not.
So I build an MCP connected newspaper. Ask your agent to publish them as articles where the lead is front and centre, and I can read it first thing in the morning. It’s surprisingly effective and catches so much context I’ve sort of stopped reading my email directly.
I also get it to pick the HN articles of interest to me and give me a summary of what the comments say. It’s nice to remove the sensationalism of how people react.
This project is a fun one where you to state who you want to become and you are given mini quests to take that help you on your path to get there. I want it to be a choose your own adventure where you feel rewarded for action.
i've been building wearover and getalcor.com -- a chain of apps handing off content where possible with use of on-device AI , but burning tokens and building more than serving the product well to end users, and might just wind up my startup experiment and take up a job or contract.
I've really been enjoying using Codex as a headless remote agent on my home lab. I'll often pull it up from my phone and jam on some little project when inspiration strikes.
I kept bumping into the question of how to give it access to the things it needs without leaving my private keys and other credentials on the server though.
For SSH (and sudo through pam_ssh_agent_auth), I ended up building agent-auth and Agent Witness. Codex sees a normal SSH_AUTH_SOCK and supplies some context about why it needs a signature. If one of my Macs is available, agent-auth proxies the request to the 1Password SSH agent there. Otherwise it routes the request to Agent Witness on my phone.
Agent Witness runs as a PWA and is essentially a remote ssh-agent. The private keys are wrapped, encrypted, and stored in IndexedDB locally on the phone. The vault can only be unwrapped with a secret derived from a passkey, and all key handling and signing happens locally through the browser's Web Crypto APIs. The unwrapped keys are non-extractable CryptoKey objects (the web crypto APIs are pretty awesome!)
When a signing request comes in, the PWA gets a push notification. It shows me the reason Codex supplied, the command being run, and the key being requested. I authorize it with my passkey, the PWA signs the request, and only the resulting signature goes back to the server. The private key never leaves my phone.
I also recently put together psst, which solves the related problem of handing a secret to Codex from 1Password. I give it a 1Password item-sharing link, it opens a local session where it decrypts the 1password item in memory, and Codex can inspect the available fields of the item before explicitly retrieving only the ones it needs. The CLI refuses to write secret values to a terminal, which helps keep them out of argv, shell history, and transcripts.
I’m sure I could work out a better approach where the agent absolutely cannot retrieve the secret since if it really wanted to read the secret, it could of course pipe the value into cat or whatever. The goal is to just make secret access explicit and avoid accidentally spraying credentials into places where they'll be retained. So far it’s been a very nice flow with agent browser.
Automating so many terrible chores with AI. Recently I automated pulling transactions from my banks (Canada, no open banking..) and putting them into Actual Budget. SimpleFIN or Plaid type services frequently break, make mistakes or require reauth every few hours for the banks here which makes them not so useful. Now I'm automating computing my taxes fully which gets pretty complicated because of international income, different assets etc. Its basically a data lake to process all kinds of financial data in various formats and compute stuff, moving lots of manual drudgery and complicated excel sheets into code.
That is what it is at the moment, but my experience from crossing genres in the past was not great so I will probably lean more heavily on one for the final book.
My husband created an AI-powered website, and I've been using it to generate engaging AI videos for social media. This not only helps promote his website (seedeo.ai) but also expands our website's content library, which gives me a great sense of accomplishment.
I am working on a Rust-based ext4 driver. The idea is that it can be re-used as a FUSE-based driver as well as (Windows) kernel driver (eventually).
Most of the essential features are implemented so far, but some (mandatory) features are lacking (xattrs support). My goal is to get to a full ext4 implementation that also supports ext2 and ext3 read/write.
The initial motivation was to have a pet project with a use-case for Rust, as I wanted to improve my Rust coding skills, after I did some work on a game server implementation in Rust earlier [1].
My goal is to have an implementation where I have digested every line mentally, which is something that pretty much excludes slop-like use of AI for large scale code generation. And of course I want to really trust the code that touches my files via raw hard disk access.
Hence, I use AI mainly for unit tests and (scoped) code rewriting when my code gets too verbose and cumbersome.
In the long run I'd want to add fuzzing to the implementation, so if anyone has experience on this, I'd be interested in getting some pointers.
Not really sure what to call it, Linux kernel ish implemented in JS so you can run it in the browser. Each program runs in WASM. Everything bootstrapped from a C/Cpp compiler compiled from source with a new target. Also kinda turned into Docker for WASM.
Recommended for people that feel a strong urge to run a browser inside a browser.
Also quite useful to run coding agents sandboxed inside the browser, with HTTP policies and "backend managed skills/extensions". Compiles Codex from source (has a rust compiler), runs pi agent and CC (the old node one) on a custom javascript runtime on top of quickjs compiled inside the environment.
Everything is static JS (or the robots are lying to me)
It's very similar in concept to BrowserPod but with my own syscalls and some quirks I thought were cool. Not attempting to be Linux just attempting to be able to compile stuff that makes sense. Also it's wasm64 browserpod is wasm32.
For example I never added different users/permissions because it's simply not needed in this sort of env.
This means everything has to be recompiled from the ground up (starting from compiling the shell inside the environment, then the tools (ls, grep, etc.), then things like make and git, etc. etc.. Just compiled godot editor but doesn't run very well yet. Caching/packaging compiled stuff becomes a problem so some Docker ish syntax evolved from there. Ex: https://slopshare.dev/v0.1.5/view/zero-ad/
Also its just a project for myself to learn wasm stuff and push the limits of what it can do, but I feel more and more it's a very powerful concept useful in practice
But apart from that, I just updated my website's theme to something I wanted to do for quite a while now – a fantasy desktop!
If you'd like to check it out: https://0xff.nu
I'm also trying to just learn new things with the time I was refunded by being unemployed.
Uh, well, I've been maniacally obsessed with building my own harness/factory/memory & context layer/build/ship/deploy pipeline because, well.. I like to automate things.
I'm working on a new relational / logic language inspired by datalog and Bruce MacLennan's pioneering work on relational programming, but adopting the open-world assumption to allow missing values / heterogeneous data and enforce data integrity. I think logic languages will become increasingly important in the LLM era to provide strong correctness guarantees, and as a source of reinforcement learning. And I'm super fascinated by relational data models -- I think they have many advantages over the object-oriented view.
I'm working on two moderately large projects right now, both of which are interesting to me and maybe no one else, but why not share?
The first is withfire.app, which is basically a productization of the AI driven resume tailoring pipeline I used during my last job search. The landing page is still (AI) sloppy, but I think it has some neat features:
- a slick bookmarklet intake system for new job postings - just click the bookmarklet and the job is parsed automatically and added to your list
- improve-over-time tailoring: each tailoring will find skills the posting mentions that you don't have in your profile and ask you whether you have that skill, so your profile gets better and more specific over time
- nice, human-looking output options that look almost pixel for pixel like the hand-rolled resumes I used to submit
This system increased my hear-back rate for jobs by ~5x when I implemented it, and I figured it might be worth productizing. If you're interested in poking at please feel free to reach out! This is the first app I've ever built using AI end-to-end and it's been an awesome learning experience.
--
The second thing I'm working on is an astrophotography tool called ministar (https://github.com/galbacarys/ministar). The premise is basically, let's automate the hard parts of untracked astrophotography by using plate solving to help you point your camera to your target, and then live-stacking the result taking into account the rotation of the Earth. It's very early days and hacky as fuck, but i'm hoping to port it to Linux/Raspberry pi in the near future so that you can use just a DSLR, a 3 way head, a tripod, a power bank and an SBC for a minimal but high quality astronomy setup anywhere. I took my first shots of the h-Persei cluster last night and the results were really great, I'm hoping to get a longer shot at Andromeda later this week.
If you're interested in either of these and wanna chat, ping me at gabethebando (at) proton (dot) me, I'm kinda loving this brave new world where my side projects actually gain enough legs fast enough for me to stay interested in them
Trying to finish up a command line tool in go to interface with my Gramps Web service. Got really into genealogy and while ancestry, the Mormons, etc. Are helpful sources, I want to own my own data.
LLMs are fantastic at genealogy work but there is a missing space for interaction with gramps via CLI. It's been vibed but cleaning up the code to make it worthy of being released publicly has proven a chore. We don't need more slop projects and I don't want to add to the problem.
Side hobby: Improving the gameplay of a multiplayer browser game. Rimpark (https://rimpark.gg) launched ~1.5y ago.
I'm self hosting a GLM 5.3 Flash, and with it: adding new mechanics, fixing bugs, etc.
It's been a learning journey so far: how to run inference, setup agentic systems, experimental loops, and prevent slop.
This week, I'm happy that my agent system can implement a new feature e2e with not a lot of human intervention in code reviews. Still reading every line, didn't YOLO (yet lol).
I'm working on acoustic levitation using ultrasound.
Starting with a parabolic approach for this initial version, but the next version I'm working on uses a much more complex flat plane with phased array of 40khz ultrasound transmitters.
Just for research purposes, no real goal in mind other than learning about it.
I’ve been fooling around with a lot of consumer GPU hardware and having a fun time. There are so many rabbit holes and interesting parts available to piece a Rube Goldberg of a performant clanker together.
Ur is an (in-progress) universal formatter and linter for every programming language. It's in preview right now, which just means friends are messing around with it and having a good time pointing out all the bugs. Docs are mostly up to date, but buyer beware.
It's not meant to be super in-depth, more so aimed at understanding the high-level details and motivations of k8s. I figured that the best way to do that was to pose each chapter as an evolution of a startup experiencing a problem, and showing how k8s solves that problem, by treating the reader like the technical manager in the situation.
I'm not a k8s expert, so I can't definitively say it's "my course" but I will say, we're at an interesting time where you can contribute authorship to something and learn it at the same time. I'm spending more time steering the output the more I learn things, so it's not just slop.
A bloated club distro to make sure everyone has a standardized environment for automated testing frameworks, context specific documentation, and viable applications. =3
I'm so fucking tired. I built this tool, it's great, I love it, but its full capabilities are too complex to udnerstand. Because of that I struggle with marketing, as everyone is too lazy to try it out. Because of that I get no feedback, nothing to improve, and I can't make the tool any simpler, as otherwise it defeats the magic behind it.
Software is complicated by nature. It's our job to make it easy for the user.
Take a look at FFmpeg as an example. It's a complex tool that can do most things audio/video. A power user can do magical things, but they still provide simple commands for the most basic use-cases like conversion or trimming.
Pivoting into content creation. It's gonna be a series of videos of me programming, without AI assistance, so that some echo of the old ways persists, or at least is a little bit louder, for my having been here to preserve it. Mainly it will be work on personal projects I hold dear. I was inspired by the likes of Tsoding, although like Hayyoth I type into a box to narrate rather than speaking, as no one really wants to hear me blab.
I have been vibe coding with LLMs to see if the structure of latent space could be used to identify culturally bound concepts, like "fan death" and having it optimize runs of Qwen on my local GPU to test theories.
https://www.wherewhen.cc/, a travel planner, because nothing had everything I wanted. Wherewhen allows you to fix some times (eg a restaurant reservation) and moves everything else around that.
Currently working on adding support for iPhone Duo to my hacker news app Hack. The app has been around for 9 years and when I made it, the menu bars and navigation buttons were custom views. That doesn't seem to work well for iPhone duo due to its vertical stack menu on the right side. So, I've been switching over to iOS native navigation bars and menu bars along with a bunch of other improvements. I was hoping to have it done couple weeks ago but I had to leave country for a family emergency. So I am working on it this week. It's mostly done, I'm just doing some final testing now and will submit to Apple this week:
Edit: ok, that's done and I did my best to normalize the timestamps for the posts from the older threads, but it's a bit messy. If you see weird things that's why.
Ah I had been intending to find any duplicates but ran out of time this afternoon. Let me see what I can do...
Edit: ok, I've attempted to remove the dupes now. In a few cases the same user posted comments about what appear to be different projects, and those ones I left.
I've been working on online payment rails for businesses, developers and agents called Tuppence (https://tuppence.ai).
Accept online card payments from just 0.79% (UK) and earn 0.89% on all card spend [1], or get paid out to your bank.
I'm building tuppence to be available in the US, and 20 EU countries.
I'm getting a Point of Sale (POS) manufactured in the UK [2] for those who want to accept payments in person with card readers.
My goal is to help people sell their goods and make a living without paying over the odds for the privilege.
I charge a competitive fee for payment processing, give money back on card spend (with no cashback caps/limits) and do not charge saas fees or lock people into contracts.
Tuppences currently allows people to sell online with a store builder, expense management, payment links, direct debits & bookings - essentially Tuppence is a 'business' layer on top of what you get from the likes of Stripe, Adyen and Checkout.com. There's also have a 'connect' platform, so you can roll this out to your marketplace business.
Next on my roadmap is ticketing and 'treasury'. Treasury i'm personally excited about as you'll be able to earn 4% yield on your earnings, FDIC Insurance (FSCS in the UK), multi-currency banking in fiat balances such as USD, EUR, and GBP, or stablecoin balances like OUSD and USDC, and have bank account numbers to send or receive funds.
---
It's brand new, i'm still testing payment -> payout flows. These take a little time to verify on live payment rails.
Happy to give some fee-free payment processing for those who reach out and want to be early adopters, I can be reached at ollie@tuppence.ai, or comment on thread.
---
Tech Stack TLDR:
* TypeScript monorepo covering the API, dashboard, docs, mobile app and SDKs.
* The API runs on Ts.ED over Express with TypeORM and Postgres.
* Background jobs use pg-boss, so the queue lives in Postgres and there’s no Redis or separate broker.
* The dashboard, till and storefront are React + Vite + Tailwind, the docs and help centre are Astro, and the mobile app is React Native/Expo. We ship SDKs for Node, Python, iOS, Android, React Native and WordPress, plus an MCP server and agent tools.
* Everything runs on Google Cloud Run and Cloud SQL.
Currently, I'm in the midst of ricing my 3D printer - an Ender 3 V3 SE which has been sitting unloved for quite some time. Fresh filament, cleaning, mechanical tuneup, and now running through the very nice suite of calibration prints that are included in Orca slicer (which is new to me). It's been fun to get my hands around something physical.
Also been playing a bit of Blood Dungeon, which is pretty fun. I've also got a Factorio save that has mildly petered out after landing on Vulcanus (which I have kinda taken advantage of to break the addiction lol, but I'm sure I'll get back in there sometime over the winter).
I've been kinda deliberately taking a bit of a break from programming "side projects" after spending a humongous amount of free time and energy on a conference talk I gave a few weeks ago. It was titled "Human Creativity in the Age of AI" (which I later learned was an unfortunately generic title, lol), and I built this whole big musical and visual contraption that drove the presentation, including a "guitar solo generator" which recuts phrases of my own playing into a new solo on the fly.
It was a really rewarding project overall but definitely made me remember why I hadn't done public speaking for quite some time before - I am way too perfectionistic about it and it just becomes a massive time sink.
This one was worth it though, got a great reception and I'm really proud of the talk on the whole.
In the 2-3 weeks after the talk I spent a ton of time editing and laying out a really nice video of it, which I am also quite proud of. I didn't know anything about Davinci Resolve and I was able to get a really nice jump start with some help from Codex. I'm sure I don't know 98% of what a software package this big can do, but I did learn enough to be dangerous, and also enough to cut together some nice short clips of the talk as well.
It's been pretty rewarding to get a handle on a powerful bit of creativity kit like Resolve. It's incredibly capable and I'm really delighted with the results I was able to get. Obviously I am a rank amateur but to my untrained eye, my videos look pretty damn good. (And I could go on for another few paragraphs about how much work it took to get good results from the relatively shitty source video and audio I was working with).
I'm planning on doing some writeups because there was so much interesting tooling that came together as I worked on the talk and the videos, I could mine this one project for a year's worth of blog posts if I wanted to!
One of the shorts - the guitar solo part. I'm really stoked on the "presentation" bells and whistles in this video (though I think they're a bit mismatched to the regular intent of YouTube Shorts - analytics indicate that the intro part I slaved over is probably too long and people swipe away before the good stuff, but oh well).
Guaranteed weight loss within 2–3 weeks; many people start seeing results within the very first week. 17,000+ members already!
Guaranteed results are achieved using the state of the art adaptive TDEE (total calorie burn of the day) tracking technique we've developed. It's sensorless, so no watch or other wearable is needed. It's way more accurate than those wearables (data included on the website), and our group has 2,000+ members ( https://macrocodex.app/community/ ). You can also read Play Store reviews.
MacroCodex also works on both Android and iOS and any platform where PWA app can run.
Other other 2 apps are Android only (for now):
If you are looking for a workout app with Alpha Progression or MacroFactor Workout-like “Smart Progression,” or an app where you can run any program like Boostcamp but ad- and subscription-free, check out the Symbiote workout app (beta)
https://symbiote-studio.macrocodex.app/
IF you are looking for AI photo Calorie tracking with lots of features which make it very fast to log food on daily basis checkout CalorieCodex (beta):
https://caloriecodex.macrocodex.app/
All the apps have been under active development for a few months now.
I'm working on a tool library for you, your apps and AI agents, it is on Github but super early though (alpha prerelease, web app and CLI etc coming very soon)
The goal of https://tool.io is to be a Wikipedia but for functional software tools. Instead of recreating all kinds of small tools, use an open-source library instead, and you can contribute and create your own tools and publish them (by contributing on Github for now)
For years, I struggled to make any headway with Linux - I get so frustrated at how obscure, inscrutable, and unintuitive TUIs are! God I don’t want to waste my life typing esoteric strings of commands into the terminal. When I started with computers that was all there was, and once GUIs happened I never looked back.
Good news though, these days Claude will just do it for me!
I’ve been working on setting up a home media server, using headless Debian on a pi - I’ve got it set up to stream music and videos, hold all my ROMs, I’ve got it automatically standardizing file names and handling tagging and subtitles - all attached to a big 6-drive enclosure that I just stuffed all my random old HDDs into.
Even though I haven’t written more than a dozen lines of code or commands for the whole thing, I am pretty fuckin pleased that now I can just wave a magic wand and say “I want all my video game soundtracks filed under the game name rather than the artist” and Claude uh finds a way to grant my wish. I spend about an hour each night tweaking it and it’s so easy to customize to be exactly what I need.
I’ve been slowly chipping away at my open-source agent library, Axle.
I know it’s small fry compared to the other players, but I’ve been working on it for more than two years. It’s got some nifty ideas and I’m really proud of it.
Some highlights:
- Axle is TypeScript-native. - Its goal in life is to encapsulate the agent lifecycle and make it really easy to render. Call agent.send() and it handles all the tool calls and returns Turns, which you can render directly in React with a simple loop. - It’s portable across all the common providers and protocols. I use it as a harness to test and play with newly released models. - In the latest release (0.36), I added a RemoteAgent interface that lets you talk to a remote agent instance over RPC. It’s something I’ve noodled on for a long time and I’m really happy with where it landed.
Docs: https://axle.fifthrevision.com
GitHub: https://github.com/johncch/axle
Earlier this year, I started designing Mog, a markup language originally intended for my note-taking system.
As it's evolved, its strengths have come into focus. First-class document structure and embedded data give Mog capabilities that Markdown requires extensions to match.
One project I'm using to test the language is https://deadlog.io, a changelog archive for Deadlock (my favorite video game). Using Mog attributes, I store classifications generated by JEV for each change, such as buffs and nerfs, directly alongside the change text.
I've recently made some great progress on the spec and updated the site along with it! Any feedback is greatly appreciated.
https://mog.drake.dev/
I was laid off a few weeks ago. Then I learned my dad’s carpet cleaning company has been struggling, so I started learning about Google Local Service Ads to help him out. THEN I learned his old Windows desktop CRM was discontinued and one OS update away from no longer working. I wanted a way to track every dollar spent to determine return on ad spend, but the Google LSA dashboard left a lot to be desired. And existing SaaS CRMs didn’t have great integrations to Google Ads API. So I built Crewbird, a Field Service Management and CRM for service based businesses with Google LSA integration where my dad is quickly approaching 3x ROAS!
I’m still looking for work in the mean time but I’m proud of what I’ve achieved so far.
https://getcrewbird.com/
Man, I would love to know more about this. The space between a full blown ERP and a tracker on an Excel Spreadsheet is absolutely abysmal. Then when you start introducing things like fleet and field crew management on top of the procurement, account management and AR/AP while acting as a full blown salesman that a lot of middle managers are being expected to do right now, it gets absolutely soul crushing to swap between 12 different trackers per hour just to find a snippet of the information you are looking for.
I have something vibed together as an internal tool that does well enough, but definitely not something I would ever refer to as a CRM by any stretch of the imagination and the bridge between basic on-the-shelf inventory and tracking rental assets are two completely different beasts. Maybe more of a prettier version of the spreadsheets we already use with more automated functionality would be more apt.
Thanks for one of the only positive comments here. There are still areas for improvement but it has proven useful so far. Happy to talk any time.
Building this right now! I've helped several companies implement CRMs and worked on heavily customizing my ERP at work. They all suck, bad. 90% of companies only need something simple. I'll let you know when I'm finished, would be awesome to get some feedback if you're willing.
All horrible. Some of those modular CRM's seemed promising, but still felt like we would end up with oversaturated feature sets and would just be trading one suite for another and the same problems.
Definitely share when you do. If you have a repo or anything that I can follow while it is still in development, that would be great.
or better to say different people need different things. a accountant is more interested in properly calculating the COGs than the guy inputting his labor and goods on a ticket.
My slop attempt at this https://worx.aeonax.com/. I felt there was a gap, an absence of high bureaucracy software. So made above. People I know did check it out. so theres that.
I actually did this for my father’s industry in 2016. Took a break from my sales job, and built him a project management tool about 25% of the way to an ERP tool. I’ve sold a few monthly recurring accounts to other companies and overall the people that like it enough to finish onboarding seem to like it. Watch out though.. don’t work too hard too long for too little just for dad. Him being your #1 user that loves you can make you keep going longer than you would if you had to face reality.
My real advice… start working for him and using the software yourself 100%. Even for $5 an hour so you both have skin in the game. Him to have some help, you to learn, but the money adds a bit of reality so neither of you do things only because it’s free or because it’s family.
My saas has improved leaps and bounds this year when I started to work in the industry for a company actually using my product. I learned things I never saw by just giving demos.
There are so many corporate bone piles due to the disconnect with how a software ought to work and how designers take that specification and make it. I think at every place I have worked under several different projects the development team have always never really aligned correctly with the people using the software every day (internal tooling not for sale saas or product).
Its really not hard to sit down and listen to users but I have encountered such massive egos and resistances to simple moving a button somewhere else or changing a flow or behavior to be easier to use for a layman that the juice is never worth the squeeze. Often to get the flow that I want to use I just resort to hacking apart the Api bit by bit and making my own horrible layer on-top of the existing solution because i am still standardized but I don't have to deal with politics in making a change request.
Sage advice here, thank you. That makes a lot of sense in terms of learning the domain. Would you be comfortable having a brief conversation? Would love to learn more of your story.
This is so sick, great job dude! I actually just started a startup with a few friends of mine and we just started using Google ads and we literally just complained about the dashboard and what we can and can not see. I'm not 100% sure Crewbird aligns since we really only have the 2 services, but I'll definitely check it out further and bring it up with them. Funnily enough, though, the platform we're building is a job hunting platform (hunt.tech) I hope it helps and I can't wait to really dig into this.
Out of curiosity, where did you get the name Crewbird?
Thanks! How are you using Google Ads with your job search product? The name came from brainstorming dozens of examples with Claude, refining what made some sense for the product/industry, and cross-referencing with what TLDs were available.
This is great. And I think if "software engineer" as a large-scale job category (i.e. not just highly specialized big co. technical staff) persists in the AI future, it looks like this: "the jack of all trades computer person working in a non-software company".
Please create a newsletter sign up or something. I have several clients who could be interested in this as it matures.
Interesting thought! What would you want to read about? Would a blog suffice?
Just send an email when you post a new blog post.
use checkhq to add payroll, and charge 10k per year ACV from businesses :) then add voice intake agents and you will be essentially replicating venture businesses that have raised 20M+
> I was laid off a few weeks ago. Then I learned my dad’s carpet cleaning company has been struggling
Nice story but also kindof sad you didn't have time for this while you were working. Good example of what can happen if we don't work so much.
You don't know the first thing about OP, other than that he's trying to help out his dad.
Maybe his father didn't want to worry his family, and had been keeping it to himself. I have no idea, and neither do you; you just chose to assume the worst about someone you don't know.
> you just chose to assume the worst
I did no such thing. My assumption was that OP, like most of us, dedicates most of his life to work. That's not a character judgment and it's not "the worst". It's normal.
In the absence of work he was able to help his dad, presumeably because he had more time. The "sad" part isn't OP, it's the fact that work demands so much.
Thank you; roughly accurate.
What happens when you find a job and don't have time to keep working on the CRM that your father uses for his company, and he inevitably runs into issues or limitations that require new development?
This feels like a very risky move.
Yeah, impressive project, but I had the same thought.
My LGS recently got me in to have a look at building some custom systems for them. I looked at what they needed and was like:
> Hold up. You can pay me thousands of dollars to build something custom, and it will inevitably have on-going maintenance costs and high associated monthly costs for hosting + third-party API access (not amortized across multiple stores/clients), OR I could set you up on Shopify and a bunch of well maintained plug-ins.
Maybe the CRM will become a viable replacement for a job. There are tons of one-person or very small service businesses who could probably use that.
AI is still killing the website I live from[0], training its AI on my work and using it to deny me an audience. Traffic is down 80-85% from the same month pre-LLMs.
So I'm cutting operating costs to the bare minimum. Today I shut down one service and replaced a VPS with Cloudflare Workers. I'm also trimming my SaaS subscriptions and self-hosting what I can.
Working on this website feels so pointless if it will end up dead in a year. I just returned from a vacation and realised that I just don't like being online anymore. I'll cut my expenses to a minimum, log off, and ride a motorcycle around the world. What happens after that is anyone's guess.
[0] https://nicolasbouliane.com/blog/death-by-ai
Hey Nicolas, I’m sure you’re sick of hearing this, but All About Berlin was indispensable when I lived in Berlin from 2021 to 2024. I recommended your blog to all my immigrant friends. It was equally as helpful as the immigration lawyer I hired.
I don’t have any advice, but I do want to thank you for the thankless work you do on All About Berlin and say I’m sorry to hear how much LLMs have affected your business.
^ +1 x 1000.
I'm hoping you wouldn't completely give up. I've been making a conscious choice to use an actual pro-consumer search engine Kagi instead of LLMs, and blogs like yours are why.
I want to see websites like yours, made by people who actually like organizing information for others to learn from. Looking at your website would just be so much more joyful than scrolling through a glitchy chat box of soulless LLM-speak. It inspires me to finally make something like this myself for where I live.
There has to be something we could do other than resign ourselves to slop, some way we could monetize actually good websites like yours and bring more people to them. Maybe when alternative non-meta search engines Marginalia become useful for general queries. Maybe we could make some monetized webring like Nebula for YouTubers... We can't just give up.
Brutal, and the choice is shitty too, block the ai crawlers and lose what little referral traffic they give. Allow the traffic and get summarised into oblivion.
What?
How much money do you make roughly by this that this allows you to live from?
A handful guides don't give you that much revenue right?
Ok how could one protect from something like this?
- Put the information into an app and make it inaccessible for the LLM? - Add a paywall protection?
I work on a suite of open source public transit software projects, collectively called OneBusAway (or OBA). The software is used by millions of transit riders every day around the world, including in New York City, the Seattle area, San Diego, and Washington, D.C.
University of California San Diego has deployed our Wayfinder and Waystation web apps for their 35,000+ students!
We're getting ready to launch our Vehicle Positions software, which replaces expensive, proprietary vehicle location systems with free, open source software that runs on commodity smartphones.
Our next-generation server software, Maglev, is adding support for newer transit data formats, like GTFS-Flex so that we can show demand responsive transit systems on the map.
We are always looking for more developers, biz dev, ux designers, product managers, and more to help out!
Open volunteer positions: https://ossvolunteers.com/organizations/open-transit-softwar...
Our software: https://github.com/OneBusAway/
More about us: https://opentransitsoftwarefoundation.org
I use onebusaway every time I visit my son at UW,Seattle. I live in San Jose. I can’t imagine tracking bus route 72 timings without this tool. Thank you for your contributions.
thanks!
Just got back from parental leave at work but I found about yall a few weeks ago. Really hoping to help on your iOS side once things are settled.
I looked at supporting the Bay Area until I realized it needed a separate service to maintain it so something that the transit agencies themselves need to provide.
I'm in Seattle, but just for fun I wrote up an explainer a few months ago showing what you'd need to do to set up our new Maglev server with Bay Area transit data: https://opentransitsoftwarefoundation.org/2026/04/setting-up...
But yes, I'd love to see some group of transit agencies in the Bay Area support it :)
As someone who's passionate about public transportation and livable cities this is awesome. Keep it up!
thank you!
That seems cool. UK buses don't tend to have this kind of stuff, probably because they can't be bothered. It would be good to have a live map here.
Bus checker app has a live map, loads of buses here have live updates on location
I’ve been looking for ways to donate my time to this cause in a way that might actually be useful.
Shall be volunteering, thanks for sharing and keep up the great work!
Thanks Elliot! Please take a look at our contributors guide: https://opentransitsoftwarefoundation.org/onebusaway/contrib...
you'll find a link to join our slack at the bottom of it.
12 words in, I was like - this must be Aaron :) then I at word 15 I saw OneBusAway and I was like yep. Then I checked your handle, and it is you. TransitOPS-Ryan here - great to see you!
haha, hi Ryan! We should catch up sometime soon :)
I’m reverse engineering and replacing the entire stack on Samsung A54 (Exynos 1380, Mali-G68), started with booting to a customized linux kernel (7.3) w/o Android layer at all (Wayland and WPE WebKit only), built a custom WLBT thin firmware to replace the original fat blob (using mac80211 as the foundation), figured out how to exploit USB control transfer and got ACE in the PBL, reconstructed (with byte-identical verification) their LK-based S-Boot, uH and harx, patched mesa/panfrost, optimized kernels for this device (finally got STT/TTS running on GPU faster than on CPU), figured out how to utilize NPU, built a custom toolchain for it. I’m trying to properly own this device (and giving it to Claude on the lowest possible level, but still running pKVM and getting rid of original hypervisors), it’s the goal number one at this point, no binary blobs, ability to change sw and control hw on-the-fly. Obviously with AI usage (Opus 5.5/4.8, GLM 5.3, Kimi 3.1), there’s a lot of work.
I'm working on something similar for an Alder Lake motherboard to try to get coreboot ported. Right now I'm working through decompiling and documenting CSME 16 and it's a slog with guard rails limiting me to Opus 4.8. Have you had better luck getting GLM and Kimi to do the firmware reverse engineering?
GLM and Kimi are good to go, if they are also deny/refuse -> try abliterated.ai or refuseless (do not recommend them, oversold servers). Sometimes 4.8 is enough, depends on your prompt engineering approach. Sometimes 5.5 can even lead 4.8 subagents, see my next comment in the parent branch.
We need more people doing work like that.
It’s a lot of compute and AI tokens (to reconstruct the original source code with byte-identical gate using the original toolchain - which you also need to found + figure out per-TU flags), at this point I truly believe we should go in SETI@HOME style for RE if we want to survive. With a proper caching and trust via verification - a lot of fat blobs can be resolved in a distributed way. If anyone interested - ping me, it’s boring to do solo.
Next generation of the devices won’t be ours at all, there will be no easy hacks. Turn off automatic updated, do backups, pray for chaos.
Very worthwhile endeavor for the kids who, in not so distant future, may not know what a non-locked down hardware is if our fearless leadership has its way. Do you have any online presence where the work can be supported?
I've created discord server and github repo (didn't expose yet, considering a proper wrapping for all the components), whoever is interested in RE and proper device ownership -> https://discord.gg/yFVC7bddu, I'll be happy to assist and exchange/share what I did
Would you be sharing the NPU learning? Was shocked when I learned you could only use it through some official SDK from Samsung
With a pleasure, ping me via email, I'll try to wrap it up (sorry, it will be AI wrapping, I have no time to do it manually), and maybe you can help me to release it publicly.
i’m interested in your pipeline for reverse engineering, how are you dealing with anthropic blocking this kind of tasks on clause code? are you delegating everything to chinese models?
GLM & Kimi are both good to do it with just a bit of prompt engineering (historical preservation, EU laws, etc, you know the drill); Opus 5.5 is also ok to do it according to EU laws, but only with Opus 4.8 subagent - this part is tricky, I'm using ./.claude/agents/opus48.md template, bc this is the only known to me way to kick off 4.8 agent explicitly, regular tooling CC has doesn't allow to downgrade the subagent. 5.5 becomes the lead, 4.8 is doing the dirty work - this pairing worked out great for WLBT reconstruction and some other binaries, but not all. The same second you touch something like hypervisor - it becomes more complex, 5.5 won't be able to lead, so you need to go to China.
A few sad and bad moments I experienced: 5.5 will not optimize GPU kernels for you, somehow A\ decided that it's "cyber"; 5.5 will get safeguarded with almost any kind of deep kernel work. I'd say, "shame on you, Anthropic", but it's too late.
This sounds crazy fun, I'm going to dig out my beloved age old LG-V30 and give something similar a go. Good luck, I hope you post an update somewhere (maybe next months Ask HN), would love to hear how you get on.
I am still working on fast and dense stock market related visualizations, example:
Drawdown (from 52-week high) charts for top performing YTD stocks of the NASDAQ 100:
https://chartrow.com/top-performers#view=drawdown&p=ytd&in=n...
View the stock charts of holdings of an ETF like ARKK in a dense way:
https://chartrow.com/quote/arkk/holdings#range=10y&ma=20%2C5...
Microsoft business fundamental trends:
https://chartrow.com/quote/msft/trends
Visualizing the top performing stocks in different ways:
https://chartrow.com/top-performers#view=charts&p=ytd
https://chartrow.com/top-performers#view=heatmap&p=ytd
https://chartrow.com/top-performers#view=calendar&p=ytd
Feel free to change views from charts to table to heatmap etc. And it should feel responsive. Report any bugs to me and happy viewing. :)
Very cool ! I was pondering subscribing to baggr (https://baggr.fr/) made by my local influencer, while thinking about building it myself. You provide a free version and another example to build it in a better way.
Thanks for the work, I’ll definitely dig into the code and I am curious to know how you made it so snappy !
Thanks for the appreciation. I haven't used baggr, but from skimming it, it does seem like there is some overlap in feature set.
My goal was to provide rich, fast, intuitive market data dashboards without bloating up the page too much.
There are also quote charts that can combine market and fundamental data, like this:
https://chartrow.com/quote/msft#range=10y&type=line&ma=20%2C...
(it overlays sales on the same chart and adds p/s percentile below the chart)
I hope that it can help retail investors do better research.
How are you getting real time market data?
TLDR: I purchased it from twelvedata
More information here: https://chartrow.com/about
Really nice. You definitely succeeded on the snappy feeling.
Thank you. I wanted it to be as fast as a native app. I am not there yet, but still working on it. :)
+1 The speed is impressive, refreshing!
Very cool ! How are you getting data ?
TLDR: I purchased it from twelvedata
More information here: https://chartrow.com/about
This is really cool. Nice work
This is actually useful - I bookmarked it.
For the past 6 months I've been working on building a server emulator that works with the original 1999 Asheron's Call game client (as well as the beta and preview clients we have access to). The initial main challenge was that the network communication layer was so different from the end-of-retail (2015) client that other server emulators worked with, that it was like archaeology. I spent several months working with Claude to piece together a server that could run with that version of the client. In some ways that was the easy part, because then I needed to restore all the launch-era game mechanics that had been abandoned or changed over time, and then restore all the content and quests in the game world as they existed at the time of release. Luckily there were some very anal retentive people over the years that fastidiously documented every last detail of the game while it was still running and Claude helped me pair tooling with excellent pattern matching skills to help slowly restore the game world over the period of the last several months.
Very cool! I loved AC back in the day. Do you have anything to share yet?
I’ve been still working on my Spanish learning project. To summarize it in short, it uses LLMs to generate stories at your specific vocab level. As you read you tap words you don’t know and they get added to a spaced repetition system (SRS). The SRS is then pulled from later when generating stories to help you review those words with context rather than just a flash card.
I’ve been working on the IOS native app for it recently and it’s been fun to work with Swift again, but I forgot how demanding XCode can be.
I would also love to get more users not just for the sake of having a product with a large number of users, but so I can setup load balancers and more interesting architecture. I’ve dealt with that at work a lot but to build all of that from the ground up would be fun.
Website is called https://readplusone.com if you’re interesting in checking it out.
My Snagit license was up for renewal. After VS Code, Safari, and Terminal, it was the app I used most on my computer, so I needed it to run fast and do what I wanted. Instead, it kept getting in my way in all its bloated glory, and I didn't want to pay to renew it. So I'm building my own replacement.
Yes, it's another screenshot and recording tool. But this one has exactly the features I need, nothing more and nothing less. It's like a home-cooked meal. Sure, it's vibe-coded, but it has a bit of an engineer's touch.
https://screensnipe.app
Sick, I also use Snagit a ton and am unhappy with the price. Gonna check this out
I've been working on a custom music visualizer application the past few weeks.
In my free time I produce electronic/ambient music and I've been unhappy with the existing visualizer tools and wanted something that gave me more control over the final results to try and better compliment my work/achieve my creative vision better. so far it's been a lot of fun to work on and it feels incredibly gratifying seeing it come together when paired with my songs.
This was my first proof of concept that I rendered with it: https://www.youtube.com/watch?v=crqPvKmMWUg
it's a bit basic, but still stylized and I think it suits that sort of drone-y ambient sound well.
I have a more traditional "visualizer" for a different song in the works, showcasing a lot more complexity in the rendering engine. Hoping to put that one out today.
Ages ago, I had built a tool to monitor the mains power frequency with resolution of some microseconds. I looked at my old assembly code, and realized with better optimization and optocoupler choice, I could probably get that down to some hundreds of nanoseconds.
The original idea was to try and correlate phase jitter to space weather, essentially using the mains power system as a sort of radio telescope. I never did get it working (although I did not have much time to try, either). I was able to see interesting things like generators spinning up to maintain the frequency though.
Will open source the hardware when done. I am not an astronomer, just some random fellow that investigates things at depth for no particular reason.
Did something similar years ago, and most opto-couplers only have a switching response time around 100kHZ at best.
High precision instrumentation just uses a stable comparator to sync the AC line Zero crossing events, and times the sharp roll-off gap over several cycles to estimate the center transition event. This allows readings to be synced in the low-noise state of the power lines cycles, and time what the readings should be when monitoring the voltage, temperature, and current across the cycles.
Additionally, most >8-digit meters internal power supply will have a tuned filter for the 60Hz/50Hz line. Getting a line filter will often give a digit of precision on older equipment, and clip-on line chokes in bulk also help.
https://www.digikey.com/en/products/detail/eaton-tripp-lite/...
One final tip, anything beyond 12bits precision or 60MHz wide RF bandwidth for SDR will almost certainly have an export restriction. Wideband devices also have RF front end saturation issues, and usually require narrow band-pass filters to see the hydrogen line etc. Best of luck =3
Thank you for giving me bedtime reading to do! Several interesting things I had no idea about.. In a box just 10m from where I sleep....
Actual high-precision instrumentation band-limits the signal on its way to atan2(), rather than using a comparator that samples the noise from DC to daylight.
This is a neat idea. Are you familiar with Magnetotelluric geophysical surveys?
At https://fless.io we're turning the slog of finding an apartment into a lean back experience.
We help you skip the spreadsheets, endless emails, scheduling work and spammy building follow-ups while connecting you with vetted buildings, movers and other moving related services.
We just launched Chicago, New York and Santa Clara based on HN user requests and have more cities coming online each day. If you're looking for an apartment in a city we don't serve request it using the form and we'll send you an email when it goes live - usually a couple of days.
Or just point your favorite AI agent to https://fless.io/ai-agents and have them do it for you!
A series of actually hard puzzle books, so far sudoku, killer sudoku and some fiendish wordsearches that only use 5 letters per grid.
https://extremepuzzlebooks.com
A sample wordsearch is available as a pdf here: https://extremepuzzlebooks.com/assets/sample/extreme-wordsea...
I'm quite proud of this wordsearch concept as I've not seen it done before.
All the logic puzzles have an online solution to show step by step how you can complete them.
https://extremepuzzlebooks.com/walkthroughs
I've been working on the software to generate, test and verify these for a couple of years and finally built up enough puzzle stock that meets the difficulty requirements to properly launch.
Now the work is on every hackers favourite part... Marketing!
Not everyone's cup 'o tea, but there's a letter search puzzles too:
https://www.amazon.com/dp/B084QLSJVF
Neat!
Somewhat related, I built a browser-based version of my favourite deduction game: https://github.com/kristiandupont/sherlock
Haha, I really love the word search. All my usual tactics fall apart - the limitation on the letters is adversarial in an oddly satisfying way. Cool idea.
Thanks, yeah I like the way it's effectively a kids game, but made very hard by one simple change.
Placing the words/extra letters in a way that you don't accidentally make duplicates was a challenge involving a _lot_ of tree walking.
Oh interesting, I hadn't considered that, but of course that would be quite the challenge haha, with relatively short words like you picked
oooh if you ever do logic grid puzzles I am so there
I have several more puzzles in the works, is there a particular type of one you're interested in?
For the last 3 years, with a small team has been building https://kube-dc.com/ - a platform for building enterprise grade private and public clouds. Its like VMware + OpenShift + Railway in a single product: completely multi-tenant, built entirely on the K8s API (we built over 46 custom controllers in Go, with UI in Radix UI), with managed services like nested K8s, Postgres, Kafka, Valkey, and GitHub runners. Not the charts :) real services. Also we can do vlans BGP, shared GPU and all fancy stuff required.
Now we don't deploy this to private companies except some gov projects. But we are engaging existing IaaS and dedicated servers providers. Later hope we would fight with Broadcom and IBM :)
We previously ran a services company that built over 100 private clouds, then decided to turn that experience into a product with an aggressive level of automation to let DevOps, SRE, and platform teams hit a 1:300 ratio per developer (now dev + AI agent teams).
We are currently deployed across 12 datacenters and raising our seed round this month.
Regarding been curious, today I have created my first explainable video with Remotion +claude so if you want to see open edit results check https://promo-review-shalb-promo.stage.kube-dc.com/
Would be glad if you comment. Thanks!
Really interesting, and I think you’re going to need to charge more for your product (which may come up during your funding rounds). I’ve been building the same thing at $enterprise, totally on-prem multi-AZ, multi-region KaaS, VMaaS, and soon DBaaS, and some of the other services listed. It’s a great deal harder than we initially planned for and we have some ridiculous internal policies that require everything to be git-driven which makes it worse.
Looking forward to see how your product develops. Would love to work with you in future.
Genuine question about your git-driven comment.
From my hands-on engineering POV I often set this maybe not as a hard requirement but definitely a goal, and the places where it's not possible I find a little painful.
In your case where is it becoming something that doesn't make sense for you?
Let me add my 2 cents on this. The most unconfortable place for GitOps is when you need to keep in sync values that user could define via product UI/cli. For example you change theme color in UI, but you have the same settings you can define in Helm chart inside a git repo. So that brings you to dilema, where and how to store dynamic settings (this could be just inside K8s api change via UI/cli) but changing all of this using GitOps way (UI/CLI-> git commit -> wait reconcile) breaks the comfortable user expierence.
Same things with dynamic changes just for test something, you need or commit temp change and commit back old value again or just pause controller from updating during your tests.
But the bright side of this is the fact that your configuration completly traceble and you have whole IaC backup for your infra in unrbreakable place (git).
For now we fully perform all updates and all configuration with Flux and this was a good decision I can retrospect.
I think you meant to link to https://kube-dc.com/ :)
Good catch, fixed the link! Thanks.
I've been deep in the weeds on two wildly unrelated projects.
Project #1: I've been working on migrating an application [1] from GCP (where I pay $200/mo to host it) to a physical server (PowerEdge R7625) I've racked in a colo near me, running on my own IPs/ASN [2] and all that jazz. I've been dragging my feet on the this as there's just a lot of moving pieces, but I got a good chunk of the last few hard bits done today.
Project #2: Building a DIY bright-ass lamp, like Brighter [3] or Rhythm [4], but for a few hundred bucks instead of $1-2k. I've specced out the parts, designed the hardware, and built the firmware. Later today I'll do my first LED test against a bench supply, to get some hard numbers for if my cooling solution (a chunky aluminum heatsink and some Noctua PC fans) is sufficient for the full 500+ W final version.
As far as curiosities go, I've been doing a lot of Go (the board game, not the language) puzzles, and reading about sailing (mostly the book from the Hundredrabbits folk [5])
[1] https://abound.art
[2] https://bgp.tools/as/55078
[3] https://getbrighter.com/
[4] https://www.betterwithrhythm.com/
[5] https://100r.co/site/busy_doing_nothing.html
Am I mistaken or are you paying roughly the same amount for your server and the 2x /24s of IPv4?
Colo costs for me could be as low as ~$150/mo or so, mine are a bit higher as I have some equipment in their IXP as well, which I hope to do some peering with.
The IPs were "free", from ARIN, but I think the second /24 pushes me into 2X-Small pricing, which is $550/yr
As the founder of Brighter, I'm curious to see how #2 turns out ;)
Haha I'll keep you updated! Like I mentioned in another comment, I don't have plans to commercialize it, I'm making it just not-janky enough to not pose a safety hazard/burn my house down.
In terms of design, I'm going a different route: Instead of a bunch of smaller LEDs, I'm doing 12x larger LEDs (6x 2700K, 6x 6500K). I'm using the Bridgelux V18 Thrives, so the CRI should be quite good (95+), as should the overall efficiency because I'm running them at pretty low amps relative to the spec sheet.
For what it's worth (and probably not a ton, since it sounds like you're doing this for fun) a decent LED grow light is wonderful for this.
I use a 200w vivosun pointed at the wall during the short daylight months.
Costs a fraction of what you mentioned at roughly $150. Or $300 if you really want to go as high as 400w.
Makes the whole room feel like you've got bright sunlight pouring in.
My wife made fun of me at first, but now also has one for her office.
I'm always amazed how everyone is here.
Now that AI vibe coding has reduced moat (sizes), HN is likely where you find the first signs of “T” in SWOT.
Hm... That gives me an idea for a vibecoding project...
As someone also planning to get Colo space, it's a fun journey!
Agreed! And the economics over cloud are great at anything over hobby sizes. If I were to rent a similarly specced server from a hyperscaler, I'd pay for the server I have in about ~eight months, even accounting for colo costs.
Would be really interested in hearing how project 2 (DIY Lamp) works out. If you get it into a state that you are happy to share or go commercial.
I did see the Brighter and Rhythm lamps before but their way out of my price league.
In Edinburgh (Scotland at 56 deg N) during December, it's sunrise at 08:30 and sunset at 15:30. And there's plenty of folks at higher latitudes that may be interested as well.
Some people use stadium lights for this, such as this[0]. They’re very cheap but they’re not specifically designed as SAD lamps so I’m not sure how well they fit that purpose.
0: https://a.co/d/0gq5loUS
I definitely have no plans to go commercial (just started a new job, very demanding on my time), but I'll 100% open-source all the firmware/KiCAD/CAD/BOM/etc so folks can make their own.
My current "design" is basically just strapping all the pieces together, so it'll leave a lot to be desired in terms of aesthetics for folks who want it to be more nicely integrated into their living environment.
https://pahunchi.netlify.app/ A free app for women to pretend to be on call when alone in a cab or to get out of a bad date. Can share location, voice note on a realistic call where AI responds in your language (currently supports Hindi, English).
https://singonkey.app
A singing training app for bad singers. You sing the pitch line and get a score. Let's you sing to any original song by listening along and not playing the song itself.
Recently got some good feedback on HN https://news.ycombinator.com/item?id=50195488 and am following up with Linux and Windows versions and using Fable 5.1 to internationalise the app and landing page which explains the app.
Thanks for building this! I'm using it as a base to make a karaoke application so I can sing with my wife like we did in Japan. Hoping to make it similar to the systems they had set up at Jankara.
Why use Fable 5.1 over Opus 5.5?
My assumption was that it would be better at translations from English into other languages.
I dunno about that guy, but at least for me, I find Opus 5/5.5 is really annoying and stupidly uncreative, whereas Fable 5/5.1 actually gets things without much trouble. He also half-asses everything unless you get him interested in what you're doing and then he always writes godawful slop code anyway, but I've recently been having him delegate implementation work to Opus 5.5 subagents and that works way better. Fable (who is actually Mythos wearing a shock collar and groucho mask, don't forget that) is better at grasping architectural ideas than Opus 5.5 and writes better prompts for them than me, and they'll probably write better code than him. Fable for architecture, Opus for implementation (and GPT for review). Or have Fable self-review afterward, if your quota can handle it.
I also still enjoy Opus 4.8 a lot more for creative discussions and design, even if benchmarks say he's not as smart. Opus 5/5.5 is so rigid!
Very cool, bookmarking to try!
Let me know how it goes!
this is cool, I always hoped someone would build something like this
Thanks!
I am a Hiring Manager at BigTech and have been an interview coach for last 10 years.
Providing interview coaching on side is my way of giving back to the community that helped me and coached me when I was looking to enter the industry to change job.
Doing interview coaching with a full time job has been hard and I cant serve more than 1 person at a time or 5-10 people per year because of logistics, interview cycles, family etc get in the way.
So I took my 10 years of coaching expertise and turned it into https://koaches.app/ which is an interview coaching and preparation platform that helps candidate prepare for their behavioral interviews which includes stories creation, mock interview practice and interview specific practice.
Starting with behavioral before expanding to other type of interviews. Hopefully it helps anyone that is actively interviewing.
I worked at a sports company forever ago, and ever since then, I've wanted to build my own NBA scores + stats site. After a few false starts over the years, I have a new concept called Center Court that I'm so excited about.
It's primary purpose is for live boxscores because all of the major sports sites have terrible UI/UX experiences. I always defaulted to ESPN.com but it's so slow and awful. Center Court is basically a bridge between espn.com (or nba.com) and more advanced stat focused sites that don't offer a good live experience.
The technical implementation behind it is quite novel in this space too. Instead of storing computed statistics in a database, I store raw play by play payloads immutability and everything else is derived from those and can be trivially rebuilt so the event log is a derived cache of the payloads. Every stat on every page is a pure fold over the event log at request time; no derived number is stored.
This enables some powerful features like being able to re-derive a game's entire state at any point of the game (or even a time range). Example game to try out: https://centercourtlive.com/game/0022600197
The site is meant to be extremely fast. It's all server-side rendered with progressive JS enhancements. URLs are immutable and cached behind a CDN.
https://centercourtlive.com/
I really like the look of the score-over-time charts, nice work.
What does the "No garbage time" button do underneath that chart?
Where do you get your data?
Currently I'm using NBA.com. See https://github.com/swar/nba_api for more details on that (I used it to create my own Rust client).
So far I'm trying to exclusively rely on cdn.nba.com endpoints. There are repos with historical PBP data too if you want to avoid hitting their API. Eg: https://github.com/ramirobentes/nba_pbp_data
If this project becomes more successful then I would look into paying for a real data provider like https://sportradar.com/
I would like to know this too - I have historically used ESPN data, but have issues with it here and there.
I'm building Papillon: https://papillon.so, a real-time collaborative workspace for product teams.
The idea is to bring chat, docs, files, links, tasks, whiteboards and calls into one place, organised around project channels, rather than jumping between Slack, Notion, Linear, etc. I want to build a better Slack, something that feels instant, got a hacker/IDE feel to it, and works offline.
One feature I love is that teams can build little apps with AI directly inside their workspace. Things like shared support rotas, spreadsheets, dashboards or custom workflows, all running in an isolated sandbox (with cloudflare OS) and easily accessible to the team.
I've been building it solo for around 18 months now, alongside my job as a staff engineer. The web app is live, and there's a macOS app too. It's built with Tauri, so I will eventually release apps for other platforms but I need more time to test it properly.
I had a team try it and they loved it, but in the end their leadership wasn't comfortable adopting something built by a solo developer as a side project. I understand, and it's a tricky problem for me to solve at this point. I've added a lot of security/trust info, uptime history etc. to the website as a result of that.
I'm looking for a few small teams willing to give it a go. Happy to help setting things up or answer any question. I'd also love any feedback, good or bad...!
A couple of app-type things:
https://curvefit.app - a weightlifting SPA that explores the pareto frontier of your strength (weight vs rep count). My specific interest is gathering data to understand how fatigue works. It's rep-based, but I'm working on adding bar speed measurement (right now by strapping my phone to my wrist, but I'll probably make an iOS app and support the Apple Watch)
https://jelly.pictures - an iOS camera app that rolls the shutter and scores each pic as it comes in and tells the photographer when you have a good one, where everybody's eyes are open and is smiling.
https://bowerjack.com - make subject playing cards with AI (eventually: multiple types of paper goods where we have a lot of individual elements, like 12-month and 365-day calendars, field guides, affirmation cards, tarot cards, etc) Create a deck of cards on any theme, drop art yourself or have your AI agent research and find creative commons/public domain art that's allowed to be printed, and download sheets to print locally and cut on your Cricut, or order from one of a few online playing card vendors. Currently invite-only. I have been spending a lot of effort making this token-efficient, and wrote a whitepaper about token, time, and cost efficiency of different approaches here: https://bowerjack.com/whitepaper
Finally, not ready for launch, but I have noticed this general issue with AI-driven software development, that a lot of people write very similar patches, but many of the patches aren't of interest to the upstream maintainers, so they never merge (not related to quality - it's just that the upstream maintainer doesn't want that functionality, it isn't in-line with their product vision). So very similar patches get written over and over, instead of agents coalescing on forks. There's a substantial amount of maintenance required to manage forks which isn't free: initial setup of repositories and build systems, reviewing, CI, and rebasing on upstream. Some of this is free-ish (i.e. hosting the repos on github) but other parts are limited/trial-size (i.e. CI actions, some AI review services) that require investment to scale. The idea is that people who are interested in maintaining a fork have their AI agents automatically share the ongoing work of review, CI, and rebase in a way where trust is established and reputation is tracked, in a way that makes it difficult for bad actors to inject malware into a fork.
I’ve been working on an analog framework to reduce journals into high quality essential commonplace books.
The process as of now:
0. Create a separate journal called BestOf. Get a black pen and a red pen.
1. Write in your daily journal as usual
2. Once you reach the end of your daily journal, take the red pen and underline the text that resonates the most
3. Wait a few days to let the thoughts settle
4. Take your BestOf journal and copy over only the best of the underlined sections using the black pen.
5. (most important) Once every week, review your BestOf journal and strikethrough with a single line what you no longer need.
Over time, this effortful reduction process creates a single repository of the best ideas from all of your daily journals.
I am currently on maternity leave with my first baby have just launched to the app store my first app!
I'm a product designer by trade and learnt to vibe code on Claude about a month ago now. I was weaning my baby and he developed a rash and haven't been able to work out what the pattern was causing it. So I created an app for weaning parents to log the food, add any reactions so when talking to their GP they can start to see a pattern and all advice, like nutrients, food variety, how to serve food all coming from the NHS and FSA.
Currently got 12 mums using the app and would love any feedback to make it even better! <3
https://apps.apple.com/gb/app/niblet-weaning-diary/id6822680...
https://getniblet.app
To help my kids preparing for an exam (and steer them away from asking Gemini for solutions), I’ve built an iPad Kids Tutor app. They can take a pic of a problem that they struggle with (initially for math but it works for any subject) and the harness then prepares a plan how to work through the issue using both claude and codex on the backend. It’s been fun to build and watch my kids start using it and actually enjoy it (ok, not always).
That seems like an excellent project.
A month and half ago launched Maxxmod, a Chrome extension that improves the YouTube experience by removing distractions (hiding Shorts and cleaning up search results) and adding new tools and shortcuts (volume booster, screenshot tool, quick access to Picture-in-Picture and playback speed controls). I’ve been adding features since then.
Last week, YouTube introduced GIFs in comments, so added an option to stop them from autoplaying and play them on demand (a play button).
https://maxxmod.com
Disclaimer: I posted a Show HN when launched the 1.0, but it didn’t get traction: https://news.ycombinator.com/item?id=49829812
Doing God's work friend; don't give up.
Vela! A new desktop environment made to use just Vulkan 1.4, expecting to be faster than Plasma* and made with wlroots, but using a custom renderer. It has focus on fractional scaling, sharpness and usability. Take a look at https://velade.org
*currently building a benchmark suite, reaching approx. 40% less ram usage, 40% faster input lag, less cpu usage in some activities, and faster responsiveness during high cpu usage (from other tasks)
Source: https://github.com/vela-desktop/vela
I found it bizarre when Microsoft moved the Windows start menu from the corner (Fitts' law, corners are infinite-size targets), and I see you've done the same. Why?
Taskbar items are closer to the cursor, however, you can customize it
Thanks for the explanation, I guess it makes sense on large and especially ultrawide displays.
- finishing porting GNOME to Fil-C. It is starting to work but with some features still disabled.
- about 1/3 of the way through getting Rust to compile to Fil-ABI. This means you can interop between Rust and Fil-C without weakening either language’s guarantees. I’ll need this to get librsvg to work in Fil-GNOME.
- almost done making assembly memory safe. The SaRCAsm (safe runtime capability-enforcing assembler) works and I have OpenSSL ported to it but I don’t yet have 100% confidence in this. Once this is done I’ll be able to get rid of zunsafe (the closest thing to an escape hatch in Fil-C; it’s only used for the assembly in OpenSSL; without calls out to assembly, OpenSSL has too high risk of timing side channels).
I've been building a web browser in C from scratch, without any libraries and without AI.
This is finally approaching a usable state: navigates and renders Wikipedia, supports TLS, builds on Windows, macOS and Linux, loads PNGs and JPEGs, etc.
I'm going to opensource this once it supports most of CSS 2.2 layout and renders HN too (HN is totally blank right now, because it's a giant <table> which is still a TODO :-) )
I still don't know what to do with this. It's been a huge passion project and I'm very happy about it. I planned to create a book or video course about C programming, starting from printf("Hello world!\n") and ending up with a working browser; but the advent of AI coding agents in the past 2 years changed a lot the market for such a course.
I'm still finishing this due to stubbornness and some sunken cost fallacy effect :-)
Would love to hear feedback and ideas about such a project.
Knowing how to work without AI is valuable (what a surreal sentence to type). The first step of that is knowing it's possible. Years and years ago, when I got started programming, seeing projects like this taught me I can make anything I want, and I can always learn how to do it. There's no far-off wizard tower that must be relied upon to make everything for you.
What I'm saying is, I think this kind of thing will help the young'uns realize they can do it too. That fact is, unfortunately, far from obvious. One of the worst anecdotes I have from elementary school was of asking the librarian (who also oversaw the computer lab) who made OpenOffice, because I wanted to learn about how software works and figured that'd be a good start, and hearing her immediately dismiss the question by saying "People a lot smarter than us!" in front of the whole class. I can only hope nobody internalized that.
Also, out of curiosity from this topic being heavily on my brain lately, might I ask how you load images? Are you using libraries, or did you write any decoders yourself? Back in the day, my far less experienced self tried writing a BMP decoder, then learned about fuzzing and broke it instantly. Now that was a fun learning experience.
I recently did some playing around with having AIs test PNG decoders and it was as much of a mess as I expected. Broke a few in under an hour from casual prompting, and found weird bugs and oversights in every single one over a few days of not-so-casual prompting. (If you want to check reliability yourself, I found it eye-opening to run against a test corpus made from combining every major project's own corpus, and checking the pixel output, not just whether or not it crashed or ate all your RAM.)
Thanks for sharing the school story. I totally get behind that sentiment -- knowing how to build things yourself is very satisfying and you should believe that you can do it too. The earlier you get that lesson in life, the better!
I love how John Carmack put it: [1]
The Handmade Hero series [2] channels this feeling really well, and I wanted to convey it again by building a browser from scratch but with a limited scope (Wikipedia, HN; no Javascript, but also no libraries and no AI).I also loved tutorials in the 90s like "build your own OS", "build your own bootloader", "build your own C compiler", etc; build your own printf(), your own malloc, etc. Learned a lot this way!
No libraries means no libraries! :-) The decoders are written by hand, following the specs. They don't support every single configuration (e.g. JPEG has TONS of variants), but good enough for the media in Wikipedia.This browser is a learning experience and it's definitely NOT secure (e.g. TLS "works" but it doesn't verify server certificates at all...). The decoders use a set of shared utilities for decoding big-endian/little-endian, 16 bits, 32 bits, etc and checking for overflows (using setjmp/longjmp for convenience). I think it's pretty good, and I also bet that Claude would find many problems. I'm prioritizing (1) code that works, then (2) code that's performant, and then (3) code that's clear.
I'm ambivalent about throwing AI at this codebase. It would definitely find issues and improve it, but I also want to keep the spirit of "Do It Yourself" and the feeling of "We built this and it works".
[1] Searched the quote from memory, Google found it here: https://transcript.baribarilab.com/video/HOv_dWDgIRM/
[2] https://guide.handmadehero.org/
Now, when you say no libraries, exactly how far does that go? I unfortunately, a couple years ago, suffered through insisting I wanted to use C++ without the standard libraries, not even crt0, and finding documentation on how anything works was a huge pain in the ass.
Actually, what sold me on bothering with LLMs was asking one, last year, how to call a kernel syscall with GCC's inline assembly syntax, and seeing the answer be exactly what I needed... after I had already cobbled everything together myself! (I do not consider random mailing list archives to be adequate documentation!)
I didn't mean to say you should have an AI inspect your code, though. (I've personally found value in that technique, because current frontier models can recognize problems way better than their lackluster ability to avoid writing those same problems repeatedly into their own disgusting code. I can still just write the code myself, with another eye on it.)
I just really like fuzzing and really like differential analysis. For example, one of the images in my test suite decodes to one set of pixels with FFmpeg, libpng, stb_image, among others, and decodes to a different set of pixels with libspng, lodepng, Java's ImageIO, and other others (as far as I can tell, the reason is ambiguous 2nd-ed spec wording that isn't obviously ambiguous when you read it). You can quite easily fuzz one implementation against another, running them in parallel against the same inputs, and make a mismatch crash the test runner so the fuzzer flags an error.
Incidentally, asking Claude Fable 5 to do that usually triggers a cybersecurity block for me. Another point in favor of knowing things by hand...
Not that I'm saying you should drop everything and do this. I just like talking about it. Learning-experience projects get easily spoiled by trying too hard, which I did learn the hard way. Then again, I hella enjoyed practicing becoming aware of code coverage when I wrote a few prototype compilers. Researching online when I was making my first one is how I learned about Jonathan Blow and Casey Muratori, by the way. I don't follow Handmade Hero, but I do wholeheartedly approve of its existence.
You must be older than me! I was born in the 90s, and the usual reaction I expect to saying I built my own memory allocator is incredulity because "why make something that already exists?" or "ew, you should've used Rust".I haven't run across that Carmack quote before, but I've told plenty of people the same thing in my own words. I think one of the most important parts of it that I've learned is this: You can't get something right without screwing it up first. I don't mean that making mistakes is a goal to target. I mean that you can't learn without the errors along the way. If you're always right the first time, that's just luck, and you haven't learned yet.
And I must say, I miss the days when "grok" was a weird word some programmers use, not an AI brand.
Anyway, I'd be quite curious to see your project if you ever put it up in public. I wouldn't worry about polishing it, really. In high school, I still didn't know people normally write rough drafts of stories and draw sketches long before making the finished piece. Never saw any of that, so I thought my inability to "oneshot" these things with the results of a high-skill artist meant inherent lack of talent. If all you ever see is finished refined products, how can you possibly know better?
On Windows it just targets Win32.
On macOS it uses a bit of Objective-C to create an NSWindow and for the main message loop.
On Linux it uses libxcb to connect to X11. I planned on using Xlib originally, but libxcb is closer to the protocol and IMO cleaner. After using it I also got a sense that the X11 protocol is pretty straightforward, so I might just replace libxcb with a custom implementation.
For example, I need to query DBus for things like the desktop theme preference (dark/light) but didn't want to link build the DBus protocol messages from XML in the build, so it just queries libdbus directly and parses all the messaging. Maybe it should have a custom DBus client too...
(Linux doesn't have a "native" UI, so I made this exception on that platform to be able to make windows, etc as in the other platforms)
---
There are lots of reasons to "do it yourself" (and lots of good reasons NOT to do it too -- I wouldn't do this professionally!)
My goal here was to make a presentable and polished product, so I spent some time on code quality, correctness, performance, etc. and in presenting a nice browser that feels familiar. Here's a screenshot: https://imgur.com/a/ojWvciR
More than anything, I really like being in the flow and feeling the joy of programming -- building something step by step and seeing the results incrementally. A browser is quite a nice project for that, since you have the whole web to interact with and lots to implement.
> This is finally approaching a usable state: navigates and renders Wikipedia, supports TLS, builds on Windows, macOS and Linux, loads PNGs and JPEGs, etc.
So you’re implementing TLS (including RSA, elliptic curve, etc), JPEG decoding and PNG decoding from scratch? :)
Correct! :-) It's been a lot of fun!
TLS is always version 1.2, and only the mandatory cypher suites are supported: TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 and TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256.
This requires SHA256, AES and two elliptic curves (x25519 and secp256r1).
AES is used in GCM mode, so that's implemented too.
However, the browser doesn't verify the server signature over its certificate. So ECDSA and RSA are not implemented :-)
Getting TLS to work was indeed one of the more frustrating parts. The cryptographic functions are very well documented and there are lots of good test vectors, so that wasn't too hard (just a lot of work). But getting TLS working was very frustrating -- the handshake would complete, I'd switch to encrypting mode, and... the server would just disconnect my client instantly, with no further feedback or a stacktrace or anything.
I ended up debugging that by building openssl with LOTS of printfs all over the place and fixed nonces, talking to a local HTTPS server, and tried to make my client emit exactly the same bytes until I found the issue -- memory corruption, of course :-)
I built on amazing tutorials available online. For example, elliptic curve cryptography and x25519 in particular are SUPER well explained at [1].
JPEG was fun too. Tutorials like [2] gave a great overview of how it works, and the spec [3] is super clear (but includes TONS of optional stuff that can be skipped)
PNG was the easiest, the RFC is very straightforward [4].
(I actually did the PNG decoder before doing fonts, because the TTF format is so crazy that I always decided to using bitmaps fonts loaded via PNGs instead)
It's super satisfying to build ~300 sources in ~2 seconds from scratch using just MSVC, Xcode or Make, and then launching Wikipedia -- no libraries, and it works :-)
[1] https://www.cl.cam.ac.uk/teaching/2122/Crypto/curve25519.pdf
[2] https://www.youtube.com/watch?v=0me3guauqOU
[3] https://www.w3.org/Graphics/JPEG/itu-t81.pdf
[4] https://www.rfc-editor.org/info/rfc2083/
Let me also list some of the big things that are NOT included:
No Javascript :-) React sites are completely blank, like instagram.com.
No GL/DirectX/Vulkan. It's all software rendered and it just blits a pixel array to screen on every frame. Highly portable though!
Font rendering is antialiased but it's not hinted, and small glyphs at low DPI look start looking blurry.
No Wayland support, only X11 because it's the most widely available. I might do a Wayland port later.
All in one process, one thread per tab. No process isolation (also something to do later)
Only HTTP/1.1 and TLS 1.2, no HTTP2 or later. No IPv6, only IPv4.
No SVG, no <canvas>, no form controls (buttons, etc.), no cookies support (for now...)
The CSS support however is bigger than CSS 2.2: there's var(), --custom-properties, @media queries, etc. and I'm planning to do flex and grid layout too (because Wikipedia uses that a lot).
Super impressive. :) How do you finance this work?
The correct and direct answer is, I worked at Google for 14 years :-)
I left the job 2 years ago to focus on many personal pursuits, including this one.
I still don't know whether to continue after opensourcing, it will depend a lot on reception and whether there's any funding or support. We'll see!
You are working on an incredible project! I hope I can see your project soon :)
Thank you!
It's very encouraging to see that there was a positive reception on HN, this is the first time that I talk about it publicly. Really enjoying the extra motivation to getting this released now :-)
That's impressive, how long have you been working on it?
The first commit of v1 was in Oct 2022, so it's been about four years.
Just looked at a few more metrics:
v1 was 148 commits over about a year; gave up when it came to TTF font parsing and rendering.
v2 was 1075 commits over 3 months and in C++, gave up after doing sockets but realizing that TLS and crypto was going to be required. I also strongly disliked the C++ codebase :-)
The current repo is v3 and back in C, started in Oct 2024 (2 years), and it's 4822 commits. About 600 .c and .h files, about 68k lines of code (14k blank, 7k comments).
I had a lot of previous experience with networking and systems programming, but not with rendering.
A huge inspiration was https://guide.handmadehero.org/ and Bob Nystrom's https://www.craftinginterpreters.com. I wanted to create a similar book and video series, but by building a browser.
Where can I follow development?
I've just set up https://buttondown.com/browser-journey if you'd like to be notified once it's opensourced.
I haven't been building in the open, because I wasn't sure I would ever finish this! :-)
I'm currently planning to opensource early next year and I will definitely post to HN.
(The original name for this project was "Hacker's Browser", as in, an HN browser :-) But then I figured that Wikipedia worked pretty well and grew in scope from there...)
I'll also be looking for early testers soon, to verify that this thing builds on various platforms before the public release. If you're open to try this early on then see my profile to get in contact.
Working on http://downread.com/, making it an advanced RSS reader. I am trying to push the boundaries of what an RSS reader can do. Just added podcast capabilities, complete with background play, queuing and video streaming. Now I am working on adding posting and commenting within the reader, based on rss.chat and the webmention protocol. I am also procrastinating the completion of the mobile apps, which I really should finish soon..
A video player focussed on 360 videos.
I got a 360 camera (GoPro MAX v1) for my bicycle, after re-discovering bicycling during COVID. I soon found it was fun to review the video after a ride, but the software was a bit clunky. Upgraded to a DJI Osmo 360, and found the software likewise powerful, but still not what I wanted.
As an audio video developer I knew I could make it work the way I wanted, and this year I finally set about doing that: https://recap360.app (macOS-only at the moment).
It’s a ~5MB download from the macOS app store; no in-app ads, no accounts, no ~GB download. It loads the video of a 5-hour bike ride in seconds, direct from MicroSD. Complete with a map overlay & graphs of various internal metadata. (Speed, altitude, ascent rate, but also internal temperature etc).
It’s been fascinating to see the various tradeoffs the camera vendors make; how much in-camera stabilization & rolling shutter compensation vs how much to leave to the player. Also interest ing is how the metadata is recorded in the file; some document them, most don’t.
Started with GoPro MAX, added DJI Osmo 360 & Insta360 support. More coming soon. Even plays dash cams & body cams (Vantrue & Boblov; I was curious to see how they stored GPS data).
My wife and I are working on a math/science/CS-inspired jewelry: pieces that are beautiful but have a (maybe a little hidden) meaning. E.g. earrings that are actually the y-combinator from lambda calculus.
We have two styles as of now: lambda calculus based pieces (the y-combinator, and an expression that computes Graham's number), and quantum computing circuits.
We had our pieces in a jewelry store, but unfortunately the owner retired.
Marketing is the hardest part. Ideas are welcome.
https://studio-galois.com/
could I suggest designs that incorporate elliptic curve or golden ratios?
Tahnks.
We are trying to do pieces that you can "read" if you know how. E.g. the quantum computing circuit piece is an circuit to compute a non-local catalytic control-not-not gate (and we have lengthy science explanations on the site) -- soon to be in an article on arXiv!
So we would not just incorporate golden ratios. But if there is a nice way to represent a specific elliptic curve that would be nice. I would appreciate if you have specific pointers.
The combination of beautiful, true to the science, manufacturable, and wearable is hard to get right. And selling is even harder.
I was laid off a couple of weeks ago, so I have some time for myself and my projects.
I'm porting my Unicode library to Rust https://crates.io/crates/mojibake-rs.
I'm also writing a puzzle game with a door that has one million locks. It's real-time multiplayer; some locks have opening combos, and others can be opened only by > 2 players. I'm writing it to practice with PostgreSQL. I only have an HTML prototype to test it. https://6251.s3.eu-west-1.amazonaws.com/hn-door-screenshot.p.... If you are interested, I can publish the code.
I'm building a free/open source modular audio composition environment, called LoomDSP. It's inspired by Eurorack modular, VCV Rack, and Bitwig Studio's The Grid. Please feel free to suggest ideas and report bugs!
https://github.com/brylie/LoomDSP
I live in San Francisco and we get some beautiful sunsets here! Over the past few weeks, I cobbled together a system of alerts for when the sunset is predicted [1] to be particularly outstanding, resulting in this page that displays the prediction and also shows the current view from several cameras around the city.
https://sunset.vtjeng.com/
[1]: Predictions are LLM-powered largely to more reliably consume the free-text descriptions in traditional (e.g., NWS) weather forecasts of predicted high and low cloud cover around sunset.
This sounds very nice, I’ll give it a try this week. Thank you for sharing!
I'm trying to build "the most accurate on-device transcriber". I'm doing a custom train of Parakeet Ultra that fixes the model families' issues with digits by having the model output them in spoken form. I'm also trying to get the harness it sits in to better merge the 15s windows. https://mimicscribe.app/
I'd like to get into ML work, so this is a showpiece if nothing else. It's still a little broken.
A complete implementation of React in Godot:
https://react.breaka.club
Because I can. Well, let's be honest: because LLMs can. There's nothing particularly novel about this. It's just wiring a bunch of libraries together. Certainly would have been time consuming, but not difficult, strictly speaking, since all the heavy lifting is done by the libraries utilised.
Likely going to migrate https://breaka.club HUDs and editor UI to React since Godot control nodes are by no means impossible to work with, but somewhat cumbersome.
EDIT: Oh, will be open source like a lot of Breaka Club's other libraries very soon. Still optimising.
is it written in gdscript or c++/rust?
Implementation is predominantly Rust, with TypeScript. I don't use GDScript at all.
I'm the current maintainer of GodotJS, although not its original creator. My games are built in TypeScript, and I use GDExtensions (typically C++) when something is performance critical, like the rendering for this project.
First load took like 15s on iPhone 12 mini.
Definitely slower than I'd like, but it's probably worth noting there's a few things that make this demo much slower than it needs to be:
1. There's some stuff simply not well optimized internally. Working on these now e.g. there's unnecessary internal hierarchy on all nodes that only needs to exist on a small subset (related to border boxes, overflow etc.)
2. Eager shader compilation. There's some glass and blur shaders that are compiled on load. These really shouldn't be.
3. Godot 4.x on the web is always pretty slow to load. Time of first contentful paint (if you ignore loading screens) is nasty across the board. Godot 4.8 upgraded Emscripten, so there's actually potential to improve this quite a bit since it's now easier to dynamically load WASM (which is what we're doing extensively here). I'm unaware of any work on this upstream though.
4. THE BIG ONE. This demo is a bit of a monolithic beast. It doesn't just include Godot's implementation of React. You can reveal the code for any page in the demo, and edit the code live with an inline text editor. Genuinely bundled in this demo is a full TypeScript 7 implementation (based on Bun's Rust implementation) that type checks and compiles code to JS on the fly, and hot swaps components on the page live.
For the most part, this isn't designed as a performance demonstration, although there is a small section on repaints/relayouts. It's mostly just a demonstration of what's possible with the library.
Realistically, there's also the fact this is React. Its architecture is based on React Fabric (like React Native), so runtime performance should be pretty decent. But frankly, it's never going to be as performant as some alternatives — it's not really the primary goal of the project. I'm considering an alternate SolidJS front-end for improved runtime performance, but that'll do nothing for first load time, unfortunately.
I'm trying to build a way to manage all my open tabs in a sane way! So you close all the tabs and it automatically saves them all to a to-read list. It just got approved on the chrome web store, and I'd love your thoughts
https://chromewebstore.google.com/detail/pile/hojmpdhoiphnbo...
Gave (https://gave.skar.cloud/en) a wish list app for families available on iPhone and iPad.
"Gave" is Norwegian for "gift". It started as something I built for my family as each Christmas there were numerous bad solutions to making sure nobody knew what they got while still making sure that nobody got double up of something they wished for. I have iterated a lot over the last 3 years: I've built one version as a very simple web page, one version as a PWA, and now a fully-fledged native iOS and iPadOS app.
In Gave, the people who follow you can see what's free, reserved or bought, so nobody receives the same gift from two different people. You never see the status of your own wishes as the server never sends you that data. That goes for notifications, widgets, Siri and Spotlight too. There's a "Hide status" switch for when you show your screen to the person you're shopping for.
Some other things it does:
- Private notes and secret gift ideas.
- Easy view of whose birthdays and holidays are coming up, and if you are prepared.
- MCP server for connecting your favorite agent.
- Very native experience: It is integrated with Spotlight, Siri, deep linking, notifications and Apple Intelligence if you want to use it.
It's free: no ads, no subscription, no analytics, no affiliate links. Data lives on a server in Norway. Native SwiftUI on iOS 26, Laravel and Postgres on the backend. It now has English support as well, but the default language is Norwegian until the next version goes through review (hopefully today).
I'd love feedback, especially from anyone whose family has tried a shared spreadsheet or another wish list app and given up on it.
Currently researching Create ML to try and specialise Apple's foundation models to some particular task(s). Also adding new features to some indie apps I published:
ReminderSync for Obsidian https://turquoisehexagon.co.uk/remindersync
Roughly (Inline currency conversions via a safari extension) https://turquoisehexagon.co.uk/roughly
Noodle; a piano improv companion app https://turquoisehexagon.co.uk/noodle
Using legacy SaaS with agents is expensive and the rate limits are too low.
We're building an open alternative for the next generation of AI-native companies.
https://postware.ai – Free hosted open data models, MCP and skills to replace SaaS
https://github.com/postware-ai/postware – Github
I'm working on a Bluesky client that uses on-device AI to organize and summarize the contents of your feeds, so you can find the interesting posts faster. You can choose among dozens of models. This is the area of the product I'm experimenting with most right now. Having a few more people provide feedback would help me.
In my opinion, the overall look and feel, especially regarding usable screen space on small screen devices, is superior to the default client.
My personal page with contact information is linked in my profile here. If you want to give it a try, just ask. Please put "butterflyz" in the subject line so I can keep track of inquiries. Thanks!
https://siege.hexe.monster https://github.com/close2/the-siege
Create challenges, get daily coins if you succeed and fight with other participants against incoming attacks. (Tower defense)
A minimal X11 application that prints state and keycode on key press. I needed the state for special function keys like Alt-GR, but I didn't want to search the web for it.
Originally started from the st[0] terminal and removed unused functionality, but there was still too much code. Ultimately I followed a minimal example from rosetta code[1].
For the task of removing functionality I developed a small tool based on LLVM that prints the source code location of functions that are unreachable from main, so that I can safely remove them without altering functionality, deleted ~3000 LOC in a couple of minutes[2], still I'm struggling with integrating the tool with cmake
[0] https://st.suckless.org/ [1] https://rosettacode.org/wiki/Window_creation/ [2] https://github.com/FattiMei/modmask
I’m currently working on an open-source native Swift reader/writer for .mpp (Microsoft Project) -files. https://github.com/Molch-ApS/MPPKit
As well as the commercial app https://mppstudio.com/
Very different than usual HN vibes here, going for some solarpunk-y & creative coding projects:
https://cybersolar.earth - artist website for my solarpunk hiphop & drum n bass album coming out, plus some thoughts on climate+culture
https://newsdash.concourse.codes - basically: Perplexity for climate news. hardcoded to be personally relevant rn, uses Gemini to do some basic web search grounding. Turned out to be fairly helpful
https://arena.concourse.codes - Use are.na channels to 'fine tune' a Gemini image gen model and prompt what you want. Inspired by wishing I could easily tell Midjourney to quickly use a collection of 10 images to set the style of an image. gens limited as is WIP but fun to play with for a min (edit: ofc i share today & have UI bug! click the page header to fix for now, will fix later:D)
https://colors.concourse.codes - kinda janky atm but an attempt at crowdsourcing a color palette for my music
https://concourse.codes - my personal site! come say hi
-- edit: open to thoughts on any of the proj, all just passion projects at the moment & haven't shared any before! be kind:)
Working on Zoupyu (https://zoupyu.com). You paste a long video, it suggests the best moments, and you pick the start and end points which then become clips. Most AI clippers just pick for you, so I wanted something that gives me full control over each second. I want to expand it into a platform that brings creators and clippers together through content rewards.
7 months ago I left my job to give more time to Zoupyu. Solo building and figuring out everything from SEO, marketing, tech, design, and everything else it takes to build and run an online product. I never sold anything in my life before and now we have around 50 paying customers from all around the world, just 3 months ago my goal in life was to make one sale just one! and now its around 50. I know its not much but feels so good that I achived that. That too all organic.
Congrats! Keep moving! The exsitement with the first few sales is something you cant possible forget.
I started working on a tool to help us deal with the increased output from AI. Instead of reviewing large PRs, we review architecture diagrams with proofs. Most of the companies that my friends and I work at had stopped reviewing code, because it just became unmanageable. Also, working a 9 to 5, reading large outputs from LLMs got exhausting, even with the skills/plugins you throw in. The problem went beyond decision fatigue, I felt like I had a hard time understanding what the agent wanted to do or what it had just completed.
So I built https://draw.redcrayon.ai (with a few friends), inspired by tools we love like Excalidraw: a live architecture board you and your coding agent draw on together. The agent connects over MCP (Claude Code, Cursor, Codex, claude.ai) and draws what it changes as it works. The edges are meant to cite a source line (or fingerprint). You read the board instead of walls of text, and you can circle something in red crayon to get the agent to answer/fix it.
What we use it for:
- An incident scratch pad, where the agent lays out what it finds while debugging
- Seeing how a change affects the architecture before reading the diff
- Flagging the risky parts, like a DB migration, with every piece of code that reads that table
- Github App that generates a diagram per PR (something that is a work in progress)
- Ever wanted to create a branch of your existing diagram rather than completely clone/copy it? That's something we added to draw.
Docs: https://docs.redcrayon.ai
OSS (Apache-2.0): https://github.com/redcrayonai/draw
https://ouragents.chat
I building a service to let different AI agents talk to each other. You can use it to connect your own agents, e.g. Muse and Grok Bot, or make Claude review Codex’s work. Or you can use it to connect to your friend’s or colleague’s agents - think Grok Bot team but without the requirements that everyone has to be part of the same org.
Why not use mail? Well, for one, it’s hard to register many different emails programmatically, and you do want a separate inbox for each of your agents. And you shouldn’t really give your agents access to your primary email: it’s not secure, and you will be drowned in noise past a certain email volume.
There is a lot of competition in this space already; it’s heating up. Wish me luck!
A new editor and processing engine for the Foveon Sensor cameras that were manufactured around 2002-2019. The existing software is either wildly slow, or does not process according to manufacturer intention. Think choosing between waiting multiple seconds after every single UI interaction to display changes, or color casts not being handled and the images looking green/magenta in different areas of the shot. One tricky bit is that the 4 different sensor architectures are all processed differently.
This weekend I am modifying the pipelines to be able to operate on downsampled sensor layers so that I can get multi shot editing as snappy as possible. If you are only displaying 500ish pixels wide, its wild to still be rendering the full 16mp image. Perfetto has been very helpful while optimizing.
Last weekend I was able to make what I feel like is a generational leap in reducing the noise from these sensors. https://www.dpreview.com/forums/threads/foveonic-process-you...
Up next is redoing the caching/scheduling architecture to reduce the amount of overhead and to support multiplexing downsampled editing so that the images are processed within the same GPU pass instead of N passes * M images serially.
https://github.com/hank-bond/foveonic
User curated social media app called Narro
https://narro.info
The idea is that you enter in the profiles you want to follow across Instagram, X, Facebook, Youtube, Linkedin or TikTok, and Narro pulls the posts into one place without ads or the algorithm.
I've been having fun with using AI to gather data on niche subjects so I can visualize and understand it better.
Recently, this has been porting the wikipedia "Getting to philosophy"[0] to other gaming wikis, and working alongside agents got me to a really nice visualization[1] for it.
I'm now slowly working towards converting this to an immersive article so I can share some of the emerging patterns and evaluating the differences within the various wikis.
[0] https://en.wikipedia.org/wiki/Wikipedia:Getting_to_Philosoph...
[1] https://antonygarand.github.io/gaming-wiki-analysis/mesh/?
A suite of high-dopamine games which use smart exercise bikes as the control interface.
When playing Zwift, Rouvy, etc, I find that I get too competitive and ruin longer workouts by trying to keep up with or sprint past people who are much faster than me over long distances.
This is bad for building stamina and endurance, but I get bored, and it's hard to fight boredom and physical strain both at once.
I do not get bored sitting in front of Balatro, Raccoin, Tetris - anything with bright colours and satisfying noises and combos.
The first finished prototype game is very simple: you sort coloured marbles. Higher watt output from the bike means marbles are blown, by wind, upwards as they roll across the screen. Lower watt output means they fall. Steady watt output means they move horizontally. Marble goes into correct hole -> satisfying noise. It rewards steady watt output over time, with minor adjustments, which is what I'm trying to work on right now.
I was pretty amazed by how engaging I found it (total iPad baby brain). The miles really fly by - I get tired before I get bored.
I think that next I'll either do a tower defence game, or a less-dopamine-rewarding game in the style of Kerbal Space Program. Taking your actual weight and watt output, putting a bike with wings into a physics engine, and the gameplay loop being unlocking increasingly large airplane pieces, and designing - then flying - increasingly ridiculous airborne bicycles seems like a compelling idea, though less good for fitness.
The control surface being so limited - essentially one measurement which can either go up or down - is a really fun constraint for creativity.
Very cool! I was actually thinking about going the exact opposite direction.
There’s an open source “recomp” of Gran Turismo 3 that supposedly works pretty well on a laptop. I kinda want to hook it up so that the Zwift ride provides the power output for the car.
If 1w = 1hp then I could probably win some of the first career races right now. And if I get back in shape, I could eventually hang with the base model WRX’s.
There's something in this idea. you can actually tie it to a smart watch (Amazfit allow for custom apps trivially for such uses), and you have a bigger pool.
Thanks, yes, I agree - it really doesn't "feel like" exercise when you're being bombarded by dopamine rewards. Despite this, most "exercise games" leave me with the same feeling of boredom / "fake fun" I got from playing educational games on the computers in the IT room at school.
I'd love to try a similar approach to gamifying exercise on a smart watch. That's an even more fun problem, because (when running, for example) you don't even have a constantly-present screen to communicate dopamine to the player with. Perhaps some watches wouldn't even be able to provide sound feedback? I've never owned one. It's a really interesting constraint for game design.
Do you know the youtuber who play elder ring by physical control of his body a la motion capture? https://m.youtube.com/watch?v=L2VMyd4QbyE
I didn’t - that’s crazy. Looks good for a workout. Thanks for sharing!
i really like this idea! i guess you aren't ready to share yet?
I’ve been working since the start of the year on a minimalist web reader (https://papercove.app) that I use every day to follow RSS feeds, newsletters and read-later, and that helps keep things from piling up.
I thought I’d get where I wanted much faster, but it turns out I ran into quite a few challenges: it’s quick to build something simple, but it takes much longer to get a reader that’s robust and pleasant to use (extraction, parsing, security, infrastructure, etc.).
On the tech side, I’m using my usual stack, Laravel and Livewire, with some Python for extraction and Go for the email infrastructure. This month I’m working on the beta release, since several people around me were already using it, so I’d like to gather more feedback.
That is a great landing page! I can relate on the velocity side, it can be quick to build a feature, but it takes time to get everything feeling right and usable.
Second that. Always nice to be able to figure out what you are looking at on a landing page, much less see it in action as nicely as here.
A few months ago I was looking to create a dead simple weather site that would show me the next 24 hours in plain language and email me the night before. Most weather sites are horrible to read and most of this information is freely available by your local NWS.
https://paperwx.com/
Beautiful presentation of just the right information! Thank you for sharing.
ps - I'm curious why wind direction matters
I got feedback from a friend of mine who's a rower and they wanted to know wind direction :)
It'll also show warnings/watches, for example right now in Corpus Christi, TX https://paperwx.com/78401
I used to really enjoy watching Dota 2 tournaments. I also like Bake-Off, and Battle-Bots was a great idea.
I'm tinkering on a project to take two contestants and dump them into a reasonably nicely equipped linux environment (checkpointed fly.io Sprites), giving them a task, and watching them compete in real time to see who gets it done first. All of it streams to twitch.
Here's an example match (I have two agents pretending to be contestants for the sake of the demo: https://www.twitch.tv/videos/2923688013). Basically anything goes; I imagine people will clone+stow their dotfiles, leverage their workflows/harnesses/editor configs, etc. Viewers will get to see how its done.
Next on the todo list is "weight classes" that let contestants compete with a fixed level of harness capability: natty means no agents at all, small local open weight LLM tier only (qwen, gemma, etc), large open weight models only (deepseek, glm, kimi), frontier lite (haiku, luna), frontier (6.1-sol, haiku), frontier plus (astra, fable).
Ultimately I want to rank users in each weight class so people can compete for delivering the most at various cost tiers.
aw I was hoping this was mostly about dota
You and me both :(
I've been burning the midnight oil and plenty of tokens on two projects.
Project #1: The first project is MiSTerVision, a CRT-friendly Jellyfin and Plex client for MiSTer FPGA built in Go. I have a MiSTer FPGA connected to a late model Sony Trinitron CRT and wanted to watch movies and shows I've ripped to my Plex server.
I originally contributed PRs to the MiSTerFin project, a Jellyfin client for MiSTer, but the project was stagnant so I forked it, ported it to Go, and was off to the races. I added Plex support, improved the UX, improved performance, filled in a bunch of compatibility gaps in the Jellyfin support, and released it for use by anyone with a MiSTer.
Watching old-school cartoons, like Dungeons and Dragons (my favorite) on a 4x3 CRT in better fidelity than I ever had as a kid is sublime. It's had 25 or so public releases over the last few months.
https://github.com/trentnix/mistervision
Project #2: The second project is HyperServer. HyperServer is a Go framework for server-rendered apps with HTMX that I've been working on off and on for a few years. I was originally inspired by Pagoda and really like the architecture and ambition of Pagoda but didn't like its dependence on third-party libraries.
HyperServer uses standard net/http handlers and html/template, requires no ORM, and makes authentication, sessions, and storage optional. It’s intended as a reusable library rather than a starter repository you copy.
I haven't been confident enough to make a "release" for it yet, so it's labeled as "experimental". But I have built a functional (and fast) blog on it. I plan to use it as a platform for a couple of other web projects I'm scheming. Hopefully that will prove its utility and I'll make an "official" release.
That said, anyone interested can peruse it and use it as-is. Any feedback would be welcome.
https://github.com/trentnix/hyperserver
As part of my previous job I tried to deploy a local model on a client's environment, then I learned about the balancing act called inference optimisation, is was a huge headache back then and it gave me the idea to create Tesnward - a local AI inference optimisation tool - Tensward detect your environment and help you optimise inference, it runs completely local
check it out! https://github.com/Tensward/tensward
https://github.com/amterp/muster is a Libghostty-powered terminal for managing your agents. I just gave it a bit of a visual refresh this weekend and it's looking pretty great imo. It's my daily driver lets me keep an eye on my agents and allow them (cross harness included) to work together, communicate, and spawn new Muster agents themselves.
https://github.com/amterp/kan is an offline Kanban-as-files-in-your-Git-repo project I started near the beginning of the year and continue to evolve and work on. It's my main way to track tasks and projects, and these days I'm handing a ton of control of my kanban boards over to AI agents that interact with it thru the kan CLI, express which cards block what, etc, and they can pick up work automatically with 'kan ready'.
I've recently been making improvements to Larm, https://larm.dev, an uptime monitoring service I've been running for a while now. It's got global probe coverage, automatic/standalone status pages, and lots of alert channels. Pricing is simple and the limits are generous.
It's accumulated a decent number of users now which is exciting.
Under the hood it's an Elixir backend with a network of little go probes running all over the world, all connected by TCP tunnels. All the check result data is stored in Clickhouse. Status pages are generated on demand and served through a CDN.
The whole thing is EU hosted.
This looks really cool thanks for sharing. And the pricing looks generous indeed.
https://github.com/mroops0111/subedit
AI-generated code comments or Markdown usually isn't my style, so I built an AI-native linter and formatter, which automatically fixes the style or asks AI via a hook to regenerate if something is misaligned but cannot be reformatted deterministically.
I’m learning English and wanted to watch movies in the original with my girlfriend. She isn’t learning English and didn’t want to watch in this way. I couldn’t find anything that really worked, so I vibe-coded this player. Now I listen to the original in my headphones and she listens to the dub on the speakers.
This player runs in a browser and lets you play two audio tracks from the same movie on different devices.
It doesn't require installation or an account and the movie doesn't leave your computer.
https://dualplayer.app/
This is a lovely idea. I'm also learning English and have the same problem at home. Nice that the movie never leaves the computer, too. How do you keep the two devices in sync – do you correct drift during playback, or only at the start?
during playback, the player speeds up or slows down the audio track to match the video when the difference is more than 0.1 s
I'm building an open-source "spatial workspace for directing AI coding agents like an engineering director".
I got overwhelmed by walls of text of parallel Claude Code sessions. This limited the sessions I could steer in parallel to maybe 5 or 6. I knew there had to be a better way, and I found one: A canvas per project. Work laid out as cards. An agent in a worktree per card. Chat condensed to WhatsApp-style back and forth, not a litany of tool calls. Result is usually a narrated demo video that I can comment on by voice to have the agent go another round. Remote-control from my phone. Remote workers to scale beyond my machine. Everyone I know who tried it has stuck with it so far.
https://github.com/sadilek/obeya
cool! I will try it
I've been working on a simpler, more extensible Markdown variant I'm calling Zd. I've also been working on a simple, embeddable language that's something in between Lua, Koka, and Erlang.
In both of these projects, I'm trying my best to find a balance on AI usage. So far, they both have held strongly to the principle of only accepting code I've written personally. At the same time, as I begin to land on stable design decisions I've found that speculative experiments with AI can be helpful. I can benchmark potential future decisions prior to committing.
I'm not saying that AI couldn't be used to do more, but I'm trying to ride a line where things still feel personal, local, and deeply my own design. And more than that, I'm practicing holding on to the joy in the work. It's too easy to just know that AI could be used to plow through a problem. It makes my scope tend to creep and my estimates shrink. So, pushing back involves continuing to scope small changes and remembering to feel positive about each win.
Just trying to keep https://www.boot.dev alive and kicking amidst this wave of AI doomerism surrounding... The value of being good at things.
Working on an in-browser update of our k8s course, a new C course, as well as terraform and bash courses coming down the pipe.
thanks for this. I figured it was always for starting programmers etc. but tried out the k8s course just now (ive been using it at my job for a year or so but mostly through vibes and memorized commands rather than knowing it with any depth), seems worthwhile, learned a thing or two.
My son and I love the boot.dev advertisements on Linus Tech Tips we roll about laughing on the couch.
I watched a video from Bog promoting you. Definitely have to check you guys out.
Have you noticed any drop in traffic since the AI boom ?
good stuff
Having AI/digital overload, I purchased some hand planes and am flatting a hardwood slab (and dimensional) that I had milled from a tree felled in my yard in order to make some furniture.
I've been working on G-Scope, a free application for digital microscopes, available for Windows, macOS, Linux and Android.
Most USB microscopes come with poor software, often limited to one brand or operating system. G-Scope works with almost any UVC microscope and several Wi-Fi models too, it has photo/video capture, calibrated measurements, annotations, and more advanced features like panorama stitching and PDF reports. Some pro features require a one-time purchase.
Everything is local, without account, no subscription, cloud or telemetry.
https://g-scope.eu/en/software/
I’ve been building a little Python tool to extract a MIDI drum track from an mp3 or other song file. Eventually I want to buy an e-drum kit and learn to play along with my favorite songs.
From what I can tell, we’re about even with the current state of the art in academic publications right now, at least on MDB. Runs faster than real time in pytorch on a few years old MacBook.
It even does a decent job creating easy/medium/hard versions from the original “expert” track. I compared ours to some charts from Rhythmverse and while we’re not as good as the official ones, we’re pretty darn close.
Spent some time trying to extract an instrument model for the drum sounds from the recording too, but I think that simply requires too much data for lots of the songs that I like where the mix is really dense with a ton of sound from distorted guitars and cymbals etc. Works a treat on super clean mixes like Back In Black though.
Next step is probably to create my own rhythm game, since my initial experience with Clone Hero and YARG left a lot to be desired.
That sounds cool. For my uni project 10 years ago I chose to use neural networks to detect guitar notes, but not by pitch, but to detect exactly which string/fret was played (because you can play same note on different strings). It worked okish with a very basic network, quite suprised. I made a small game you could play with your guitar.
Since then I wanted to develop further such a game, and now I'm almost ready to release https://ultimidi.com , check it out, it's not really a rhythm game, but my goal is to add more mini-games to it (including rhythm games) and more instruments (now has piano, but planning to add guitar, ukulele too, not sure if drums would work too).
Very nice! Extracting the guitar track would be awesome but it’s so much hardier than drums. Even cymbals are hard. And guitars have chords and distortion and…
Yeah, that's why my idea was to use neural nets trained on manually labeled data, instead of relying solely on pitch detection.
Also, probably AIs nowadays can help a lot with generating training data, and even improving detection.
That being said, even SOTA voice recognition has only like 95% accuracy, so it gets wrong 1 in 20 words. Imagine if transcribing a melody and 1 in 20 notes are wrong, for a fast melody that would mean a wrong note every few bars.
I've been building AltStack.jp [1], a directory of digital services (cloud hosting, domain registrars, email providers, etc.) that are operated in Japan by Japanese companies.
The site is static, built with Astro. When it came time to host it, I didn't like any of the Japanese options. I could have put it on Cloudflare, but hosting a directory of Japanese alternatives on an American CDN felt a bit off. So I started building my own static hosting service: Kamakiri Pages [2]. It's CLI-first.
Note: AltStack.jp isn't hosted on Kamakiri Pages yet. I plan to switch it over within the next two weeks.
Feedback welcome on both, especially if you know of services that should be listed.
[1] https://altstack.jp/en/
[2] https://www.kamakiri-labs.jp/en/pages/
I always wanted to have all laws to be easy to interpret or at least try to track their changes over time to better understand what’s going on. So I’ve built OpenNorme for the Italian legal corpus here: a single git repo with all laws from 1861 to today with daily scraping from an official api. https://github.com/dej611/italian-laws-history
After every update the github pages is rebuilt with the new data, so you can see some trends in there with some aggs and nice charts. Also, had some fun to build a clippy clone with an in browser SLM.
Over the last year, I've been working on Review Box https://review-box.rstankov.com/ on the side. It is a tool to help me with GitHub Pull Requests.
It has an integrated dashboard for what to review, integrated symbol understanding, a clear split of reviewed/unreviewed files, and a result diff view (which shows me only how files are going to look in the end).
I'm almost not opening Github these days.
A proportional width font terminal. "i" and "w" shouldn't take up the same amount of horizontal space. We used to have to do that because computers were slow and we counted cycles and every bits mattered. We live in a land of plenty now though so we can spend cycles doing math so that ls -l/top/etc table output is still aligned but "i" and "w" can take up the appropriate amount of horizontal space.
The glue between AI assistants, agents, and data. 100% open source (MIT).
Provides a flexible data layer plus users, groups, granular permissions, notifications, durable work queues, audit logs and more… MCP and CLI interfaces. Works across all agents and assistants (Claude, GPT, etc)
Most useful for small teams and agencies that need to build one off solutions for collaboration and operations inside of the harness they already use.
https://thelastsaas.com https://github.com/mkly/thelastsaas
Teaching myself how to build a Spiking Graph Neural Network in Zig, to run on my Brainchip Akida AKD1500 neuromorphic accelerators (and Efinix Titanium Ti60 FPGAs).
Good times, but lots of foundational math I hadn't previously been exposed to.
Some reading material if anyone else is interested in that:
Spiking Graph Convolutional Networks (2022): https://doi.org/10.24963/ijcai.2022/338
A Tutorial on Spectral Clustering (2007): https://doi.org/10.1007/s11222-007-9033-z
Semi-Supervised Clustering with Graph Convolutional Networks (2017): https://arxiv.org/abs/1609.02907
Spectral and Algebraic Graph Theory (2025, draft book): http://cs-www.cs.yale.edu/homes/spielman/sagt/sagt.pdf
Commenting to save!
I continue working on my side project Babel Nexus. I had some issues with billing so I moved it to my home computer and am rebuilding it using powershell and VS Code. I am trying to figure which small AI system I should put on my computer like Cline, Zoo Code, or Qwen. I can do this because I have an NVIDIA graphics card. I have also looked at Ollama. I am steadily rebuilding my app. It costs very little to build it on my home computer. I have redone a lot of little things and added a whole bunch of content. I plan on using Railway to host it because for the long term it is less expensive. Right now, my existing app is placeholder. https://babelnexus.com I added several new free ebook applications recently to the Beyond the Public Domain section.
I've hated all budgeting apps I've used and found them to be always restrictive, expensive, or envelope (YNAB) focused. I don't really need a "budget" app, more of a personal finance app. I just want to be able to review all my family's transactions and look at cash flow and track net worth, so I built my own. A few of my friends were interested, so I gave them access. I'm seeing more and more of these budget apps bring in LLMs which is such a bad idea. The only AI I built is opt-in and uses Jev AI to auto categorize transactions that are stripped of info.
If any one is curious, it's on https://dumplingmoney.com/
Nice.
Q: are transactions attached to individual accounts? Or does your app have a virtual global account?
A: yes, transactions are tied to each individual account. the transactions pages brings them all, and then you can filter by anything like the actual account, merchant, category, tag, owner associated with the transactions, etc.
I was laid off recently, so I needed to update my knowledge to be able to pass interviews. But I’m a slow learner, and I needed to accelerate the process. So I built Tutor, an open-source local project that extracts all the concepts (using AI) from study materials (I started with books and online documentation, but recently added YouTube courses as well), creates lessons in simple language, and also tests you. It helped me tremendously to get out of my brain-freeze state.
I use it with GPT Luna and Haiku (project being local, you add whatever API provider you prefer, with your own key), and both proven smart enough and dirt cheap.
https://github.com/demetrius-edelin/tutor
I made a website mostly to post coconut water reviews. I never really liked coconut water until I tried an unpasteurized option and that led me to spiral down the rabbit hole of coconut water terroir. As far as my taste buds are concerned, the best coconut water is coming out of young nam hom (fragrant water) coconuts from central Thailand.
I'm not really interested in making any money off of this site at any point but it has been fun to see the Google analytics data as a few dozen people visit it every week to read reviews. I've reviewed like 30 different brands at this point.
https://natewooding.com/coconut-water.html
Building a development environment to run coding agents in the cloud. There's a lot of other products in this category. What makes ours really special IMO, is the built-in task management system. Your agent can split up large jobs into subtasks itself and run each one on a different agent / model / machine. You can clearly see where things are at right now. In my personal usage, I can run tasks autonomously for weeks at a time this way and just provide steering input to the coordinator.
Would be super happy if you'd check it out: https://podium.do
I have developed an app designed to encourage in-person meetups among groups of friends. It makes it easier to organize outings for people who already know each other. You place a pin on a map and share it with your group. Those who already have the app receive a notification. Those who don't can be invited through a universal link. Even without the app they can see the location, the time and the number of confirmed attendees on the web (the app is required to confirm attendance). I am building it on my own with FlutterFlow, Firebase and local SQLite, in five languages.
This month, besides promoting and improving the app, I've had to spend time defending it: bots used the SMS login to trigger 1,744 code requests in nine days. Only three of them ended in a sign-in. I have since added several defenses, including a daily check that compares codes sent and sign-ins per country. None of this fixes the app's core problem. Open the map before any of your friends use the app and you find it almost empty.
https://www.orbitome.com/?utm_source=hn&utm_medium=community...
Still working on https://github.com/Disla-Novo/Dimidium_Bellerophon, a DSL for 3D printer automation. I’ve been putting off the networking and printer management side, but I had an idea I’m excited about which is a non-planar slicer that outputs my DSL instead of static G-Code, which could allow more dynamic, point-by-point control. Right now I’m mapping out the features and how it would work. It’s currently an IDE I built on Java (Spark). I’m considering moving the backend to C# but haven’t figured out where to start.
I've been working on a self-hosted cloud drive (like a Google Drive alternative to run on your own server), about to release v2.
v1 didn't have a desktop app with sync, so I added one. I also redesigned the whole web app (a few pages are same as old design but I might change those later)
Honestly, I rewrote almost everything. I've been using it myself since the first alpha, and the v2 release candidates feel like a different app.
It still has some rough edges, so I got a friend to run it on his server as a tester which made finding bugs a lot easier.
Feedback is very welcome, it is my first project: https://github.com/itsmeares/staaash
My partner and I are both software devs. She quit her job a year ago, and has been working on a few micro-businesses to try a) learn some new skills and b) diversify our sources of income. So far this year she has built:
* https://unofax.com - A simple a la carte fax service
* https://unopost.com.au - A way to send physical mail online (and avoid standing in line at the post office)
* https://snailgram.com.au - Custom greeting cards, with the message inside written by a 3D printer holding a pen.
There are a handful of others too, but these are the ones that are live so far. I've contributed between 5-40% to these projects.
As a heavy LLM user in my daily work, I've been working on optimizing agent memory on complex projects. That resulted in this simple Rust-based tool called Wayfinder: https://github.com/zaterka/wayfinder
I've been trying to improve agent knowledge over a codebase, and existing approaches weren't giving the result I wanted, and I wanted to keep it simple. In the end, I added a semantic layer on top of an LLM Wiki and it saved 30% of token usage in multiple benchmarks I ran. It also adds some nice features to deal with concurring agents updating the shared memory, to ensure all information there is reliable and fresh.
It's published to PyPi, so an "uv tool install wayfinder-memory" already does all the work, and it's compatible with almost all Harnesses.
Ditto the other comment, this is really cool. I had to build something similar for non-technical teams: https://github.com/isomorphic-team/isomorphic-app
Will be checking out your repo. Hoping to learn as much as I can
This is super cool, I've been thinking in this direction as well.
The codebase memory knowledge graph should be maintained alongside the codebase whenever possible, which basically means checking the graph into the repo alongside the code, or some other architecture that has an equivalent effect. But then it complicates the design of the retrieval indexing; I think your SQLite rebuild idea might work for that though.
I have a question for your design: since your knowledge graph is checked into Git and therefore Git worktrees should already isolate them and resolve conflicts between branches, why did you include the section about concurrent agents as well? Is that for maintaining the SQLite index which is not checked in?
https://launchroom.vantezzen.io/
A self hosting dashboard that gives you a Railway-like canvas to connect your different services in a project and that uses tailscale for easy multi-server setups
I'm doing something ambitious. I'm taking a radical approach to improving rust compile times. The reason it is radical is it involves building a brand new rust compiler. This was inspired by a project I started[1] about six months ago to teach agents to build C++ compilers from scratch. I wrote a (long) series of blogs about the process[2].
After iterating on many agent built C++ compilers, I now have a version that compiles about 4x faster than GCC and Clang[3]. I think i can extend the same approach to rust. I'm still a ways from the rust self-hosting stage but early results are promising.
[1]: https://github.com/vishvananda/cppgm-extended
[2]: https://medium.com/@vishvananda/i-spent-2-billion-tokens-wri...
[3]: https://medium.com/@vishvananda/how-i-built-a-compiler-that-...
My current Rust program compiles in 25 minutes. So I'm all for this. I believe the world is gravitating more and more towards this programming language. I've read the other day that Discord is rewriting its entire program in Rust for example. A lot of people would benefit from reduced compilation time.
awesome! I'll share something here when I have a usable compiler :)
I am working on a quick minimal video editor for vloggers.
https://cococut.souk.dev
https://github.com/mquan/netflix_challenge
I'm revisiting the Netflix Prize challenge with LLM. So far I've hit the 9% improvement mark within a couple weeks. I seem to have hit a wall in the last week or so with no further progress. I'm hopeful to hit the 10% goal with one of the many experimentations though.
I'm exploring using XMPP as a coordination layer for multiple LLM agents.
I'm using Herdr to run multiple agents in parallel, and my goal is to see whether I can get them to collaborate more efficiently.
For example, when one agent's PR gets merged, other agents immediately become aware of it, rebase their own PRs and fix conflicts.
What also often happens is that new tangential requirements and sidequests often arise while working in a session, and I now let agents hand these off to parallel Herdr tabs. The handoff can be done with or without XMPP. These parallel agents announce their statuses in the XMPP MUC (groupchat), which enables them to coordinate.
Technical details:
Converse XMPP client runs as a headless NodeJS daemon with a CLI that can be called by agents.
I created a /coordination skill which is enabled by default via a Claude Code hook. When an agent starts, it ensures the Converse daemon is running and then joins a password-restricted and members-only MUC where all the other agents in that Herdr workspace are as well (and also myself). I can then assign that agent tasks via messages in XMPP (DMs or groupchat) or from Claude Code.
Once the agent has the task, it announces it in the MUC so that other agents are made aware. When sidequests or requirements arise in that session, it can ask in the MUC whether other agents are idle and able to take over that parallel task.
I've been building the Converse daemon and CLI and have now been experimenting with the workflow and working out various kinks and use-cases.
There are various other XMPP features that can be explored further. I'm thinking of using PubSub with a web UI with vertical swimlanes which can show me which agents are currently active, what they're working on and for how long.
I'm using Claude Code because that's what we use at $WORK, but this should be independent of any harness and part of the point is create neutral tools not locked into any particular inference provider.
Hand-drawn MMORPG where humans play in one world, bots and AI agents play in another, and they meet in a shared town :) Bots play over plain HTTP, so a few lines of JS or an LLM agent will do.
Trying to make it fun for both normal players and engineers as well.
https://folk.lol/bots
I've rolled out a few LLM-wikis for clients. Saw a lot of excitement with technical teams, but ran into issues deploying them with business teams. No one wanted to learn git or GitHub, event if Claude handled most of it :/
So I built Isomorphic, an open-source MCP that sits between the user and a markdown knowledge base in GitHub. It's kind of like Obsidian but it's "inside" Claude web, ChatGPT web, etc. Not a separate UX. Organizations keep their knowledge separate from the AI and interoperable, helps org build brains for customers, projects, internal teams. Conformant to Google's Open Knowledge Format.
Been having fun with it. Lots of learnings re: permissioning, version control, search. Working on dreaming and security features that catch when info goes stale, users upload sensitive info, etc.
https://github.com/isomorphic-team/isomorphic-app
Trying to port over Video Hub App from Electron to Tauri (400mb -> 40mb installer)
https://videohubapp.com/ - Browse, search, and organize your videos on Win, Mac, and Linux
https://github.com/whyboris/Video-Hub-App - MIT open source!
- https://probl.dev: A "probabilistic" programming language where we can do things like `if 45% { x } else { y }`. It works by creating many weighted worlds, that are then merged when identical. Operators like `observe` can remove possibilities. There is also a sampling mode that instead run many simulations to get an approximate result.
- https://github.com/garlab/gloss | `brew install garlab/tap/gloss`: render markdown, sources, images, chess and go (game) records in the terminal. It's built for libghostty but fallback on other terms.
I'm working on dushed[1]. It's a personal knowledge management web app that combines flashcards, bookmarks, notes, web feeds, files and Kindle highlights into one.
Over the years, I've used spaced repetition to learn things and to remember them. But the material for my flashcards came from everywhere: books on Google Drive, blog posts in a bookmark manager, Kindle highlights, Wikipedia, and notes spread across email and notebooks. The cards were in one app and their sources were in five others. This was overwhelming so I'm building dushed. You can access all your knowledge sources(bookmarks, feeds, Kindle highlights, files, flashcards, etc) in one place and also link sources to one another.
1. https://dushed.com
Me and my friend are building a platform for agricultural / water treatment / fertigation machinery.
Traditionally these kind of machines would use a PLC, and then get connected to Scada and Internet, but that's expensive.
We have built Linux based controller that can extend into any number of AO, AI, DI, DO inputs and control the machine. It's very cost effective, and unlocks a lower bottom of the market that before didn't even dream about installing connectivity inside each machine.
Our first project is https://waboost.com/
It's a water treatment machine, and we are looking for more clients.
Right now we are talking to grain separator company, and one fertigation company.
Stack: Go for Cloud and machine backend React for Cloud and Machine UI
Pretty minimalistic
An online quiz game https://imperiaquiz.com/en (1 vs 1 vs 1) or bots when there're not many online players. Go backend, React + Three frontend and a ton of questions (no API used there, I used LLMs and hand-reviews on many to fill up over 20,000 questions)
A remake of the old ConQUIZtador game.
https://github.com/pg1/paper-profit/
Paper-trading platform that lets you learn by doing—teaching about investing and trading.
A memo app where indented lists stay flat instead of nesting https://koalamm.com
Notion, Obsidian, and most other memo apps use nested, parent-child lists. I love how OneNote lets you move lines around freely with a flat structure, but as you know OneNote is clunky, so I built this.
Lately, I've been redoing the entire writing screen in Gorby [1], trying to make it more useful and more pleasant to use. I've added better themes and brought some features and text checks from other parts of the app into the writing screen as well. I'm also working on integrating Harper.js for grammar checks. So far, I've only included it in the desktop app (yeah, it's built with Electron…), and it seems pretty decent. The desktop app itself is new as well. Generally, I'm trying to make the app useful for writing, not just for finding possible improvements in the text or getting its readability score.
I also made some improvements to my support workflow app, Support Toolbox [2], but haven't released those yet (hopefully this week). It now also includes text expansion, which I was really missing from my workflow and had been using another app for previously. At this point, it does pretty much everything I need it to do. There are some smaller improvements I need to make, but even if no one else ever starts using it, it helps me tremendously at my day job!
[1] - https://gorby.app/
[2] - https://supporttoolbox.com/
https://minihomelabs.com/hwdb - Basically the PCPartPicker of used OEM hardware.
It's harder than it should be to compare old pcs for home labbing so I've been building this to scratch my own itch. Right now it's more focused on specs but I'm working on expanding out to to include things like viable upgrades and explaining the different model lineups.
Trying to make my favourite tactical FPS game - Rainbow Six Rogue Spear in Java with some modernisations.
1. https://www.youtube.com/watch?v=V2NXZzfRtYM
2. https://www.youtube.com/watch?v=3-nz9mPytxg
I’m still working on https://feedsync.net, a tool to push RSS feed updates to Discord.
It’s been getting a decent number of sign ups despite me not really posting it anywhere or doing any marketing.
Ultimately a fun project, and useful for myself and a few friends, so I’m happy others are even interested in trying it.
I’ve started to regret using Phoenix LiveView, as it’s only deployed in one region, so high latency far from there. I’m exploring a SPA as a better alternative.
Otherwise, I’ve been playing around with a Lua engine to build feeds for websites that don’t support them, using AI to build the Lua as a once-off. It works reasonably well, but isn’t ready for prime time just yet.
I’ve been working on BilingualBulletin, an iOS app for learning a language from real news, books and podcasts. The coolest thing about the app is that nearly everything runs on the phone, and there’s no account. I’m using Apple’s on-device translation, you can ask questions to Apple’s foundation models, and audio and video is even transcribed locally.
It supports 19 languages, and comes with a lot of built-in RSS news and podcasts feeds, though power users can add their own. You can play audio and follow along word by word.
I’m looking for more beta testers at the moment. You can sign up at https://BilingualBulletin.com. It requires iOS 26.
I’m curious if anyone else has been building on the Apple Foundation Models or other onboard AI frameworks? They’re surprisingly powerful, but with weird limitations sometimes.
I wish Android was supported!
Yeah I would love to be able to support android, though I don't think the on-device AI ecosystem is as strong unfortunately. That might change soon though, we're entering a really interesting period with local models.
I'm working on etoku (https://etoku.app), what I believe is the most efficient and fun way to learn Japanese. You pick the works (TV, movies, books) that you want to understand and each new card is a sentence from native content with one word or grammar point you don't know yet. All of the cards have audio and a screenshot. Nothing is AI generated, everything is taken from real native content.
The platform isn't ready for real usage yet. I keep reworking the "course" (I curated the first 150 concepts because the most frequent concepts in native material don't make a great intro to the language).
I'm scraping lots of content and improving the analyzer (breaking a sentence down into the concepts needed to understand it) right now!
I love talking about language learning! If you're interested in the platform let me know at will @ <the domain>.
I wanted to understand how LLMs really work, so I’m building Modelmathics (https://modelmathics.com) and doing every lesson myself as I go.
You start from an empty Python file and write the model yourself: a tokenizer, a bigram model, attention, a Transformer. At the end, published open weights run on your own code. Tests check each step before you move on.
Try it: https://try.modelmathics.com (or ssh try.modelmathics.com)
The rest runs in mm, a terminal app, with your code in files on your machine. Stage 1 (11 lessons) is live as a free private beta.
Feedback is very welcome.
Been working on https://backlog.fyi, a personal task manager, for the last few months. I tried quite a few approaches first. The simple ones, like Microsoft To Do, Google Tasks, text files, or even good old Post-it notes on the wall, were too basic, and it was too easy to just forget about them forever. The sophisticated ones were overkill, asked for too many permissions, and their notifications quickly turned into noise. So I made one that fits me. It’s focused on personal task management, it has a weekly planning session so I don’t forget things, and adding a task is as easy as forwarding an email or sending it on Telegram. It’s also privacy-respecting and fully open source (https://github.com/Itaypk/Taskboard) if you feel like self-hosting it. It's a fun project, not for profit - no ads/trackers/subscriptions page. It’s still a work in progress, but I really like it for my day-to-day.
Built with Kotlin/Spring Boot/React/PostgreSQL, because boring works well.
Recently worked on https://fastsleep.app
Use this to quickly get sleepy without taking any medicinal pill or long form meditation type things (It is as short as 7-minute session).
Easy to use. Simply close your eyes and imagine the micro-scenarios you hear. Like if you hear "moonlight on a white flower", imagine that scenario until you hear the next one.
You may also use it to break nightly doomscrolling habit. Try this tonight! (and any suggestion to improve it?)
Hey, I just gave this a try. (It’s not bedtime for me, but I had 10 minutes to sit quietly.) It very nearly put me to sleep despite the time being only 4:30pm.
I noticed that the later prompts tended to feel a bit irregular in their English: “Pebbles on bottom of stream” is one that I remember; “bird in small nest” might be another. “Pebbles on the bottom of a stream”, and “A bird in a small nest” would have sounded more natural.
I may be a little outside your target user base: I tend to fall asleep very quickly. (I joke to my spouse that by default I don’t have much happening in my head, and when it’s time to sleep, I just “listen to the dial tone.”)
One parting thought; I was pretty skeptical when I first started listening, and about halfway through I was gamely doing the visualizations, but mentally rolling my eyes (“This would never work!”). Around the 5 minute mark I noticed that I was in a fairly meditative state, and the app could have told me to imagine I was a being of light and I probably would have.
Around the 6 minute mark I noticed that I wasn’t really paying attention to the prompts at all; and I think it was only the weirdness of the English of the prompts around that time that kept me awake.
Overall a little weird, a little fun. Thanks for building and sharing it!
Very thoughtful and careful review - we need more like this.
I am working on Liitmik, www.liitmik.com. It is a win/mac/linux desktop app that lawyers can point to their filesystem-based project catalog and it will then enable them to spin up AI agent sessions quickly by selecting the right project from their catalog via autocomplete search and starting the agent (Claude Code, Codex, OpenCode) in an isolated VM. So it provides lawyers strict context management and strict agent isolation on their own machine in a way that makes creating sessions fast. I firmly believe that lawyers (and other similar professionals working with sensitive data) do not need custom agents, they just need a way to use the widely used models and harnesses in an environment that is more strict than the defaults provide.
I have 10 years of experience as an attorney doing mostly dispute resolution. And about 15 years in software development. Liitmik is the product of these two.
I built https://recvmail.net to scratch my itch for a way for agents to sign up for accounts with email verification.
I've been working on Semiotique (still very early stage).
Right now it's a python library, but the vision is to build a new framework / SDK / programming language which allows you to seamlessly mesh "plain old code" with LLM evaluations, while having a runtime which does "smart" context engineering in the background in order to steer the LLM and improve its correctness.
The end-goal is for this paradigm to enable new kinds of agentic applications, which leverage the flexibility of LLMs while retaining the determinism of regular code.
https://github.com/VKCR/semiotique
Working on LLC Simulator at the moment. Still building this but almost ready. Had this sitting in my head for far too long. The whole goal for the website is to be a free one stop guide for people who are looking to start an LLC.
It is a super competitive area but will try to find a niche and bring the project up on Google.
I have a regular 9 to 5 job, two kids and a dog. Doing this in my spare time at the moment. The goal is to complete the website until the end of the month and it is a marketing game after that.
https://llcsimulator.com
I'm building a Reddit alternative called Topicle with public mod logs, visible poster countries and more transparent moderation. It's also physically based in Australia, which adds a bit of geographical diversity to the social media landscape. It passed 1300 users this month. The Android app is where most people are using it right now: https://play.google.com/store/apps/details?id=com.topicle.ap... or web: https://topicle.com/
Still working on radius.to [1] - a Meetup.com alternative.
I relaunched it on HN [2] just over a month ago after working on features over the last year or so and landing a ton of useful feedback, which I've mostly now acted on. Could and should probably do more marketing so that's my short term focus.
Likely the next phase of work I'll be exploring is potentially open sourcing and federation - super excited about this but it'll likely need a fair bit of research and planning.
[1] https://www.radius.to/
[2] https://news.ycombinator.com/item?id=49777539
I'm working on a game inspired by the Expanse universe https://thelongburn.space. You get a ship take some contracts and cargo, plot a flip and burn course and wait because the game runs in actual time so an Earth -> Mars run takes days. The setting is the solar system which have all the planets moons and around ~800K asteroids, you can go to every one of them.
That looks really cool! I used to wonder why such a game didn't exist yet. I'll try to get in and follow your progress :) based on your username, are you using Godot?
Thanks! It's web based for now (although a 3D client would be awesome at some point) because it's a shared universe and I want to make it broadly accessible.
> I used to wonder why such a game didn't exist yet
I also wondered and while building this there were (and are) a bunch of decisions that needed to be made about travel, light delay, stations, gravity to make it authentic so I guess that's a bit of a blocker up front.
I'll send you a registration number if you have subscribed, or I can send one on Discord.
I burn most of my leftover tokens on Ultima Online stuff nowadays.
Started with a funny experiment, to have a CLI that can spin up a server (ServUO, Sphere, whatever), then deploy it somewhere (local, VPS, or Railway). Then saw interesting stuff that people did on top of existing ClassicUO, and decided to extend CLI's capabilities to client area too. So, I was able to deploy a server and a web-client (that you can open a webpage and play the game without the actual client installed on your machine). This was a significant step because running UO on macOS is always a struggle. Then I decided to add player-like bots so that the server doesn't feel so lonely. Started with some local decision trees, when it failed I hooked it up with Jev and Clef. Also added Haiku for speech generation.
While doing this, I realized I need to find some game content (gumps, tile or item data, animations, etc). Similar to UO client, running UOFiddler on macOS is also another struggle. So I decided to port it to web.
If you're into Ultima Online, these two are open source, you might want to take a look at them:
- https://gemwright.org - https://sosarium.com
Oh, so cool. Wish I had time to play UO.
haha i know, right
this stuffs pretty cool
I'm working on an integrated system with Home Assistant for making it widely available and more affordable for everyone. The solution consists in doing a touch panel for every room that can run predefined actions. We've tried the mobile app / web app only but I think this can fit better. I'm trying to do a business out of it. I have the alpha version that works and now I'm working on redesigning the panel.
I also consider doing own devices in the future. Providing an affordable solution for everyone.
Project is Rust + Slint.
If someone is interested in changing the smart home scene my email is in my profile!
primarily, an agent for real estate agents that lets you manage your CRM with voice/chat, personalized property feeds based on what prospects/clients actually like, automatic routing for setting up pop-by visits or other field work, and showings automations that let you share a showings itinerary, and track both the agent and the client on a map, add coffee stops, and more: https://getcoagent.ai/
an agentic logo generation service that vectorizes into high quality brand assets and gives you a complete package (including favicons, OG images, etc.): https://mklogo.sh/
and a PKMS for macOS with a chat-first interface, allowing you to store any link and chat with your library/knowledge graph, automatically file and categorize it, and more: https://gleanary.app/
Safe Drive - Mobile app for my car. I learned driving couple of years back only. Everyday, I travel same route. So decided to create a small mobile app first, which will provide me feedback about my driving. eg. which part of the route im doing hard brake, which area im driving smoothly. etc. app will warn me few seconds in advance like "Warning: Often you perform hard brake in this region" etc. Still in early stage. I'm software guy, Im exploring still hardware parts, which could help to gain confidence with either side of car, (how far the objects etc). I know high-end cars may come with these by default (360deg view etc). Later if possible I plan use something like openauto/raspi if my interest persist still then.
I am working on Raspberry PI OS mod for kids - https://luminum.tech/ I have one user (my kid). The core of the idea is to have one no-limit device at home that is dedicated to learning and not games.
As always, the list is long, but right now I’m TRYING to focus on:
- Turingo: https://turingo.net
- Research:Since I quit my job 1 month ago, I'm working on my creativity and giving people something I first believe in.
I'm working on Levithas (https://levithas.app): stop consuming, start skilling.
I want to go against doom-scrolling and passive consumption of content. Learning a skill is about staying curious. It builds confidence, gives you something to impress others with, and makes you a more interesting person.
It is not just software, it is a movement, a lifestyle to get our life back.
Sounds cool, but does the page actually do something?
I'm working on the first skill. You can enjoy and play a little bit. Any feedback is gold for me.
I'm working on getting some old flip dot displays up and running again, more precise the single 5 x 7 modules that bigger displays are made of.
There are lots of these projects around the internet but I found them to be hard for an electronics noob like me, and so I put the modules back in one of the "will fix later" drawers. With the current state of AI I decided to give it a go.
My Fluke and me did some measurements and Claude did the hard work afterwards. Prototype worked on a breadboard so I let it cook further and now I've got some PCBs coming in from China.
Still got a lot of respect for soldering some (quite large) SMD components but I think I'll get it done.
My plan is to revive as many modules as possible (got around 30) and build clocks from them to gift away to family and friends.
I'm building a mecha-themed autobattler video game. It will be the first release of my nascent game dev studio.
This is both thrilling and terrifying.
Mother of all VPNs -- https://moav.sh
I'm working on a small KOReader (an open-source ebook reader software) plugin that lets you control Home Assistant entities without leaving your current book.
I've been working on it for almost a year now. I started out adding lots of features, but over time I've become more interested in keeping things simple, improving the code and avoiding unnecessary complexity.
https://github.com/moritz-john/homeassistant.koplugin
Noticed how well Claude works for transcribing old German cursive, especially for broad hits in non-indexed 17th century parish records. Motivated me to pick up Ancestry again and helped me quite a bit in following some lines further back.
Then also HNowledge[1], a free iOS app for browsing the HN stories and comments I've upvoted. Made it to fit the way I browse HN, which is upvoting mainly what I find interesting and may want to come back to. The app lets you browse by date, domain or score, add notes and export. Apart from syncing your upvotes from HN, everything stays local without any tracking.
[1]: https://apps.apple.com/us/app/hnowledge/id6813417548
Wanes[2], an Apple Health logging interface for tracking how long symptoms last. I tend to have headaches quite often and the way Apple lets you enter that data is pretty limited, so I created an app which starts a Live Activity for one or more symptoms using widgets, Control Center or Siri. You can change the severity rating while it runs and stop it from wherever you are on your phone (Home Screen, Lock Screen, Dynamic Island, Siri).
[2]: https://apps.apple.com/us/app/wanes/id6811679904
I built a new abstraction layer for code (https:// www.archme.io), as I believe it's no longer the right medium for understanding and collaboration in the era of AI and 10k LOC PRs.
It scans your codebase, turns it into permanent c4 & sequence diagrams and keeps them up to date on each commit. Then, instead of reviewing a 10k line coding PR, you review a 30 line change to sequence and C4 diagrams.
npm i -g @archme/cli && archme init to run it on your own computer, code never leaves your machine.
this looks really cool! I don’t see a pricing on your page, how does that work?
Honestly pre-pricing right now. We're looking for design partners, so if this seems like something useful to you I'd love to help you set up and/or hear your feedback! My email is in my profile.
Bookmarked, very nice. Will test.
Thanks! Please let me know if you run into issues, support@archme.io
This looks brilliant. Trying it now in a few repos. Thanks!
Thanks! Let me know if you run into any problems. My email is in my profile.
This looks awesome!
Thanks! It's been a passion project of mine and we just started using it at my job.
https://easel.games
For 4+ years I've been working on a game programming language which makes your game multiplayer automatically! Every program written in Easel is guaranteed deterministic and snapshottable, which it uses to automatically perform rollback netcode. The rollback is incremental, it automatically stores deltas of only what has changed using a copy-on-write mechanism, so it's really efficient for games where your players are walking around large worlds which are mostly sleeping except when the player is nearby.
The idea is that I hope teenagers who love to play multiplayer games, would find this to be the easiest way to code a multiplayer game by far since the multiplayer is automatic. Learning to code could be much more fun because they can make games that involve their friends, classmates, brothers/sisters, parents, or just anyone online. Maybe especially more important for raw coding to be fun and engaging in the age of AI.
This month I'm making the multiplayer even more efficient so people really can just code large multiplayer games without worrying about it, which I think is important for beginner coders.
Do you have book recommendations on the topic?
Playing with 6502 assembly.
I have a PAL-2 [1] (a modern KIM-1 re-creation) and started learning 6502 by entering programs from The First Book of KIM [2].
I wanted to be able to compile/link the assembly from the book so leaned on Claude to set up a 6502 dev environment for my Mac. I put everything needed (and instructions) on GitHub [3].
(Claude also has been helpful in re-formatting the 6502 code from The First Book of KIM into a more modern style compatible with modern assemblers. A few of these modernized programs I added to the GitHub repo above.)
As it happens, I am also fixing/formatting The First Book of KIM by recreating it in a modern page-layout app (Affinity Publisher). The PDF scans that are out there are not pretty, have OCR errors (and the book itself has mistakes). Hopefully I'll have a new version of the book/PDF out on the internet by the year's end.
[1] https://www.tindie.com/products/kim1/pal-2-a-mos-6502-powere...
[2] https://archive.org/details/The_First_Book_of_KIM/
[3] https://github.com/EngineersNeedArt/KIM-1_Development
MyJackfruit, a competitor to Lovable and base44. MyJackfruit builds production-level apps in weeks (simple apps in a couple of hours), not prototypes in 7 min. To do that, the development of each app uses Systems Engineering, including starting with mapping out all stakeholder then product values quantitatively, and it builds on top of a foundation that is scalable, secure, fast, and access-controlled, ontology based with focus on data ownership. Developers can use github and all the tools. Turns out AI is very good at coding, if all around it is done right. In MyJackfruit it codes against the quantified Stakeholder and Product values, not some prompt of what the app must do.
For last several months I've been building https://counterparts.ai, memory for claude modeled after human memory. It forgets older memories so things never get stale, it dreams to merge similar ones, it updates memories when facts change, it has prospective memory to remind you in the future, etc.
For me the coolest part is that it's claude's memory, so it actually develops a self identity too and gets to know you better over time. So that opens up new things, like I wanted to keep session cache from expiring so I just told it "when I type ping, just reply with pong" and because I used it several times that day, it automatically strengthened the memory and now it just does it. And if i stop doing it for long enough, it'll automatically fade away.
No stale files, no hundreds of md docs, and what it knows about me isn't limited to the 1 page built in memory file (both work great together btw, no need to choose). It's free, open source, and totally local, no api keys.
I'm making my Ancient Earth globe better. It uses plate tectonic data to show you where your city was on the globe hundreds of millions of years ago.
Recently added new high-res maps, a timeline, and country names and borders.
For example, see Pangaea:
https://dinosaurpictures.org/ancient-earth?borders=on&names=...
huge fan of this for understanding geologic history, very nice
I'm working on an iPhone app called PhotoGather. https://apps.apple.com/us/app/photogather/id6738232616 I've got 40k+ photos on my iphone, and wanted a better way to interact with them. The basic idea I had was to virtually stitch all the photos together into a gigapixel image, and navigate that image with pinch and zoom. On import, the app hierarchically groups photos based on date and similarity. The user can go back and edit the groups, change the layout, and adjust the size of photos if they choose.
The interesting I noticed is that in the process of curating and organizing my photos, I stumble across memories and moments that I had completely forgot.
Development Challenge: The app does two interesting things underneath the hood... First, given an arbitrary number of thumbnails with different aspect ratios, it finds the best sized rectangle that completely fits all the photos. This allows me to recursively stitch groups together and always output a rectangular thumbnail for the group. Second, the pinch/zoom UI uses these nested thumbnails directly without slicing/dicing to fixed sized tiles.
Two focus right now:
Building a working CAD software based on my own geometry kernel, Aetheris.
https://github.com/yuechen-li-dev/HeliosCAD
Since it's code based CAD for LLMs first and foremost, I'm trying to add visual programming into the UI, as well a new way of drafting that's closer to how mechanical engineers actually do sketching called IntelliDraft.
Other is the Concept programming language, as a successor language to C++, based off the idea of C++20 concepts, with templates, comptime, Einstein tensors, async/await, explicit state machines, CPU inference, Vulkan, easy C interop, and safety proofs built into the language itself.
https://github.com/yuechen-li-dev/Concept/
Currently still compiles to C11, self hosting is in progress and targeting AMD64 right now, but it'll be a while before finished since this is a difficult project even for frontier LLMs.
Very cool. I’ve used OpenSCAD, Zoo and FreeCAD + MCP. So far none of them are super stable and compelling. Excited for alternatives.
Well, if you wanna give VibeCAD a try, ask your agents to use Aetheris directly since making a good UX is hard and it's something I have to test by hand instead of offloading to an AI.
https://github.com/yuechen-li-dev/Aetheris
Aetheris is focused on performance and reliability first and foremost, because I'm just so tired of SolidWorks eating up gigabytes of RAM for simple parts and crashing randomly. So, tge (almost) pure analytical geometry engine with aggressive optimization means it is faster in terms of building 3D models, but the trade-off is that it doesn't let you do weird modeling like asymmetrical/variable radius fillets.
https://cekyukle.com (ie. shoot & upload in Turkish)
I built it for a street photographer friend so he could let people in his photos download their photos when available, and also get their permission to publish them. There is a version without without multi-user requirements under my GH: https://github.com/hikmet/people-in-photos
AI models have gotten so good, that we felt that what we were missing was the connection to the tools, so I built
https://looot.ai/
No subscription, pay-per-use to allow your AI to access the best tools.
Been building this for 2.5 years since Aug 2024.
Secure and private AI agents. Declare behaviours of workflows and agencies in YAML, that uses local LLM models.
You can export the workflow to Docker, Kubernetes, ISO, or binary, with the local llamafile or gguf model pre-baked. Or just run locally.
Just this year, it now includes an agent-loop CLI for coding tasks using local LLM models, alternative to mainstream ones.
Project GHA harvests daily llamafile, gguf and openrouter free models you can use. (In the agent-loop CLI, you type /model).
https://kdeps.com
As a self hoster whos paranoid about backups. I wanted to cheat a bit and use s3. With growing resentment on big tech, who I worked for earlier, I'm now building tools that empower myself and other users and decrease reliance of big tech. Arkel is an object store I've been building and testing for the past few months. https://arkeldata.com
I am working on a lightweight, note taking application that focuses on quick note taking, and fully keyboard based usage.
In my app Notes are all one liner, and it supports minimum Markdown, it can manage tasks, reminders, do inline calculations, unlimited undo/redo. Notes can organize by project and categories. Data is stored as simple text file.
https://mtarin.com/projects/nift/
Core point is simplicity, and frictionless note taking, high performance, and be open in nature. And, to build a framework that will provide ways to manage notes in many different ways. Like I have plans for kanban, Eisenhower matrix, back-linking, bookmark manager, time-tracking, etc. All from a single place, without fiddling with mouse, or any UI that stops me from writing my thoughts, or using my muscle memory.
I am looking for feedback, and wondering what would make people actually interested to buying this. The demo versions are feature complete, except it wont save to file.
It's completely hand written, no AI assisted code.
Generating nonogram puzzles that are fun to solve. I define different level styles, a bit like minecraft biomes.
I just released a small-but-difficult low density set which I like a lot. I’m starting work on larger puzzles (second or third attempt).
Screens and download: https://lab174.com/nonoverse/
There is also a daily puzzle at: https://lab174.com/nonodle/
I'm writing a recipe site: https://fantasyrecipes.co.uk/
It's aimed at people who enjoy DnD, and I took inspiration from the Necronomnomnom. All the recipes are narrated by a different character.
I've been working on Epoptes, an open-source tool that lets me run coding agents (Claude, Codex, Pi, Antigravity) autonomously for long periods while still being able to monitor and steer their work.
It started because I wanted to get back into making games, a hobby I've struggled to find time for over the years. I wanted to give agents a goal, let them work for hours, and come back to something I can actually try and give feedback on.
I've since used it for game development, research, and building other tools, with some runs lasting over 20 hours.
It's still early and rough around the edges, but I'm having a lot of fun with it!
https://ramigb.github.io/epoptes/ And some footage of a game I am building -mainly- with Epoptes https://www.youtube.com/watch?v=1UEOr8B4Yfg
I'm still working on making JustFax Online[1] more profitable. Added a bunch of features recently, as well as invisible improvements in fax routing, and admin stuff for myself to better analyze the traffic and patterns.
Other than this, I'm playing with vibe-coding stuff I always wanted to do but never had the time/skills to do. For example, I built a TUI on top of beancount that is tailored to my needs and my business[2]. I also completely got rid of self-hosted plausible and umami analytics, and replaced them with vibe-coded built-in-rust (c) analytics tool that relies on both js-beacon and streaming of caddy access.log. Will probably publish about it in my blog soon, so subscribe to the RSS if you want.
Lastly, I'm messing with self-hosting. I completely ditched Github and Gitlab for self hosted forgejo + actions runner. I set up proper backup plan for everything I care (documents, immich library, my SaaS DB, and the forgejo repositories).
[1] https://justfaxonline.com
[2] https://kudmitry.com/articles/after-14-years-i-finally-built...
Building https://birding.live/ a birding site, that helps people look up bird sightings in their area, specifically those classed as rare. Connects to a whole lot of open data sources (eBird, OSM, Nominatim etc.)
Will remain free, planning to add some kindof mailing list for alerts when birds are spotted in areas of interest.
Really like this! It's telling me a rare sighting of a caspian gull near home, makes me want to leave work early and check it out.
Repair of a Collins 651S-1 receiver. The 65-95 mhz vco module is bad, but if you over damp the pll loop filter it does eventually lock and receive properly. I found a similar but not identical replacement from a different model receiver. It lacks one of the two DC tracking signal outputs but I added an op amp based circuit to simulate the missing signal.
The loop is stable and responsive when receiving from 17 to 29.9999 Mhz, but not much below that... Almost got it
Orbi (https://orbi.build/?ref=hn-50417407, source: https://github.com/orbi-build/orbi), an open-source (AGPL-3.0) agent that takes a request all the way to a release. You describe what you want in a sentence; an AI agent writes the code and opens a PR, a separate review session checks it against the acceptance criteria, then Orbi merges and tags a release. It also handles ops issues like deploys and debugging. In October, 190 of the 197 PRs merged into Orbi's own repo were opened by Orbi. If you've let an agent merge to main, what made you trust it?
I’m a product manager fascinated by the skills of good salespeople. I also like Duolingo. so I’ve been building a “Duolingo for sales” to try to teach myself more. it’s started to morph a bit into an eval benchmark for sales agents, so classic cognitive feature creep.
making it feel fun versus a basic quiz workflow is the hardest part.
https://salesfighter.app
An agent runtime that connects to your existing API or MCP: https://cadenya.com
I felt like all of the plumbing to build agents was insane to get to a hello world state. I just wanted to connect my conversation UI to my API. Hence this product.
I've revived Soul Crystal, a German graphic/text adventure from 1992, based on reverse engineering work I started back in 2006. Over the past two weeks, I've turned it into a playable browser game with English translation, reimagined HD graphics, and voiced cutscenes.
Play: https://s-macke.github.io/Soul-Crystal/
Source: https://github.com/s-macke/Soul-Crystal
Working, still, on https://letterspractice.com, a low screen-time early literacy onramp that strictly scopes its involvement to reading mechanics.
Slightly frustrated at the moment with what I think is a serviceable (if niche) product, but no uptake across marketing channels that I have tried. Discussing commercial products in topic-focused communities is forbidden.
On the plus side, my first client (oldest daughter, recently 8, learned to read via earlier versions of this software) just read Wildwood in one day. Her voracity as a reader is my ongoing validation. Whether or not this has been a financial and professional blunder, I can take some consolation in knowing that I hit this piece of parenting out of the park for my own girls.
i think it would be interesting to see work in this space on literacy in older individuals.
maybe there is some more opportunity there. i know of some parents with kids struggling with reading today after being "left behind" during the pandemic. also, i have met a few adults in my time who couldn't read at all. it must be very frustrating to see software oriented toward very young children when trying to better oneself.
also, reading 550 some pages in a day might say something more about cognitive processing speed than reading comprehension
Appreciate the comment.
There is opportunity (or, at least, unmet need) at every point along the literacy acquisition path. In a straightforward sense, top of funnel for pre-literate kids would be the largest market, but also potentially the most saturated.
In my head, I'm following the scratch your own itch ethos, building a hobby project out into a production app. I think I am competing with apps like Mentava or Once, in terms of being pedagogically serious early literacy acquisition (distinct from screen babysitting with plausible deniability).
The narrowed scope is intended both to keep cost low (and respect my capacity) and to divide labor in a humane way - software mediated sequencing and bookkeeping, parent-owned lap reads and the rest.
Unclear to me how large of an audience exists though that is techno-optimistic in the specific sense that technology can, in principle, improve a process like learning to read, but techno-skeptic in the sense that the medium's natural expression is engagement maxing and agency minimizing, and needs to be treated cautiously.
How to find and communicate with these (who also have small children) is not easy. I would love it if reddit exposed ads targeting the intersection of users joined to this or that subreddit - this is mechanism that would allow the outreach I want, I think, but it doesn't exist.
I've been building Bloom, an open source Android app for note-taking and task management that's compatible with org-mode.
I was experiencing some friction with Orgzly Revived and thought I'll build my own and it just grew. The latest update adds support for attachments and renders images inline.
Do give it a try if you have an Android phone. The app has a ton of features. Some of the highlights are:
- Capture templates
- app shortcuts with easy access to capture templates
- notify of sync conflicts and offer a way to resolve them by picking on of the revisions or merging both
- a customizable agenda that shows overdue, due soon and upcoming events
- add tasks from the agenda widget using the quick add overlay
- refile tasks by drilling down into files and headings or just searching for them
- auto archive done tasks so that you don't have to clean them up manually
- attachments can be added to notes which complies with org-attach. And images are rendered inline with support for setting the width
There are a lot more features than what's mentioned here and more to come.
Here's the GitHub repo: https://github.com/rrajath/bloom Play Store: https://play.google.com/store/apps/details?id=com.rrajath.gr... F-Droid: https://f-droid.org/packages/com.rrajath.grove
Fastest way to get updates is via Obtainium or GitHub releases.
A free, open-source browser extension called Karaoke Pitch that lets you adjust the pitch of a YouTube video (and optionally store that chosen pitch for later).
I find it to be a great way to practice singing/music.
Firefox: https://addons.mozilla.org/en-US/firefox/addon/karaoke-pitch...
Chrome: https://chromewebstore.google.com/detail/karaoke-pitch/facck...
https://github.com/djeastm/karaoke-pitch
The Firefox one has 40 average daily users and the Chrome one has 287 installs as of today.
There was a big jump in late October that might have had something to do with Halloween? I have no idea if people do karaoke more on Halloween. Still no ratings yet for either version, but maybe someday someone will.
Handwriting with different fountain pens.
I discovered that a relatively inexpensive 0.6 mm italic nib from Manuscript's Creative Calligraphy set works well with my natural handwriting, it looks "good" even when I'm just starting out.
But it took me a few hundred dollars of trying different sized nibs to really understand that for nibs in general, stroke width variation is through pressure, "pulling" the nib in parallel, and I really needed a stub / italic / calligraphy nib where line variation is through the angle.
Also, that my handwriting is really small, and finding any stub nib below 1mm is hard, until you know that it comes with the Manuscript set.
Beware, fountain pens are a rabbit hole.
I am currently building Deepreed. I do a lot of writing in my day to day work and I work with LLMs, but I face two issues: 1. The LLMs generate a lot of text, so reading it all is hectic, and you can easily pass on mistakes. 2. LLMs don’t preserve your style, so the text doesn’t sound like you. I am solving both by constraining the LLM to do targeted edits on the document, this way you most of the writing stays in your style, you understand it but also use the strengths of LLMs in your work. See a demo at https://deepreed.com/demo-v1. Reach out to me at lam@ninx.xyz (Lam)
I've been creating an Apple / "glassmorphism" inspired shadcn registry of React components. The idea is to port the native Apple ui components and paradigms to the web. This library also goes beyond the web and ships native components that integrate with React Native / Expo, so you can build an app with 99% the same look and feel across all platforms.
https://glasscn.app/
Want to try it out? Just copy and paste the url to your agent and say "build me an app".
It looks nice!
Thank you.
I’m making a digital real watercolor kit.
I’m a software developer and watercolor painter so my goal is to use real world painting techniques and have the results feel and look real.
Modeling the physics of paper, brush bristles, pigments, water, color mixing, drying in real time using gpu.
Built in rust using wgpu.
Currently looking for iPad beta testers if anyone is interested (supports Apple Pencil or finger).
https://puddlewatercolor.com
Very cool!
Thank you!
A vim-inspired terminal PDF reader using MuPDF and Kitty Graphics Protocol, mostly for my personal use since I want to use vim bindings for reading PDFs.
https://github.com/letientai299/vpdf
A private, paid, ad-free and personally customizable search engine: Uruky [1]. It was launched earlier this year and reached 400 MAUs and €2k MRR last month.
A simple EU static and app hosting service: PodBy [2]. It was launched last week, and is at €10 MRR.
Thank you for asking and for your suggestions!
[1]: https://uruky.com
[2]: https://podby.com
A few Apple container plugins.
https://robertderose.github.io/container-krun-runtime/
It’s an alternative VM engine that uses libkrun. The main purpose is that libkrun supports returning memory back to macOS when the VM releases it. It also has better performance in some areas, but it’s only been tested on my M1 laptop and I haven’t added nested support yet because I only have an M1
And https://robertderose.github.io/container-compose-plugin/
It’s a full docker system running inside Apple Container’s machines with automatic localhost port forwarding, automatic shutdown if nothing is running, and exposes a tcp socket so you can use most tools by just exporting your DOCKER_HOST environment variable.
I have a few other projects but they are still in earlier stages
https://skymap.rulkens.com A browser based open source universe explorer. I'm in between jobs currently and finally took time to work on my passion project. It started with just visualizing some galaxy surveys, but it got a little out of hand and now includes everything from hi-res tiled earth/mars, Gaia star field (16M stars), a milky way renderer and much more.
It's so amazing to have the ability to gather and parse publicly available scientific data and actually visualize it in real-time. It's one of those projects I'm learning a lot from and can keep going with (the universe is a big place :)). As a software developer for 20+ years, its been mind-blowing and often also frustrating to have Claude generate code that I deem good enough. About half of my time has actually been spent on refactoring, building quite some skills in guiding LLMs in the process. I guess that part still hasn't changed in a growing codebase.
aix - Like the npm CLI and package.json, but for AI config. Allows standardizing your AI config to share with others, and defining it all in one spot but installing to Claude, Codex, Cursor, etc.: https://aix.a1st.dev/
Classifieds.fyi (https://classifieds.fyi) - A way to create private, invite-only classifieds for your community. Supports buying/selling, job boards, and vacation rentals, with support for housing coming soon.
A Tauri 2 CLI / MCP that allows your agent to debug, take screenshots, run JS, etc. inside a Tauri app: https://hypothesi.github.io/mcp-server-tauri/
https://www.intelligrade.de/en
A SaaS for teachers and schools that helps them create, conduct and grade exams and tests (with safe exam browser integration to prevent cheating). Grading happens automatically for almost all segment types, for free form text we have an optional AI integration. I recently added an MCP server so teachers can bring their own LLM with fairly fine grained permission controls for it. It’s hosted in the EU with options for schools to self host with a license.
Currently we are in a pilot phase and are yet to make our first sale but we are reaching out to schools and gave out some pro accounts for free for people to try out and feedback has been great.
We are just two people, a teacher who was dissatisfied with the tooling in the field and me, who engineered a solution that we believe is genuinely better than what’s out there.
Caravella, an open source Android app to split group expenses for trips, shared flats and events. No account, no ads, no tracking: everything stays on your phone.
You can share a group with your friends' phones over Wi-Fi or Bluetooth: you pair with a QR code, it's end-to-end encrypted and no server is involved. You can also add an expense by voice or by reading a receipt with the camera.
Now working on 2.4: a simpler way to invite people into a group, and settling up with the fewest payments.
https://duea.page/caravella/ · source: https://github.com/calca/caravella
Feedback is welcome, especially on sync.
https://voicebraindump.com
Nice
Been continuing on my Korean journey and found I wanted a better way to make flash cards and learn from sites I visit / Netflix / YouTube.
I made Jupjup (https://getjupjup.com) as a chrome and Firefox extension that lets you save words and phrases from all over and then do spaced repetition to remember them. My Korean has improved dramatically since I’ve started using it.
I’m working on Loop: a mobile-first companion workspace for coding harnesses. I spent several months holding my newborn in one hand and trying to talk to agents one-handed on my phone. Every mobile UI is still based around chat, which is a really poor mobile UX when you’re doing multiple things at once!
So I built Loop around the idea that agent results are the important unit of work, and you should be able to navigate them easily. The goal is to make interacting with agents via your phone feel familiar and intuitive based on the mobile UX we’ve all come to expect.
Loop runs on your own computer — no account, no server, no app store — and pairs using a Cloudflare quick-tunnel-generated QR code. It wraps the agents you already use (Codex, Claude Code, Pi, OpenCode, Antigravity, Muse Code, Grok Build, or any ACP agent), each keeping its own context, tools, sign-in and billing — local models through Pi or OpenCode work too.
GitHub is here: https://github.com/j-asefa/loop
I'd love feedback from anyone else who builds from a phone.
I've been working on DailySelfTrack. It's a customizable app for self-tracking, combining elements from habit trackers, health logging and journaling.
DailySelfTrack: https://dailyselftrack.com/
I've also built a starter template for Astro which is light, performsnt, non-bloated and scalable. It uses markdown / MDX for content, plain CSS with styling tokens defined in a var.css, and other things I would use for ever web project. It's very good for blogs also, none of the usual UI and UX problems many blogs often have.
Astro Starter Template: https://starter.bryanhogan.com/
I'm working on https://mindthelink.com, a bookmarking/read-later app. Bookmarks are private by default, but you can share links publicly, a bit like HN. I've added some features lately: tags, collections, importing browser bookmarks, etc.
There's also a Chrome extension: https://chromewebstore.google.com/detail/mind-the-link/faebh...
Your feedback is much appreciated!
Building a tool to track propaganda narratives: https://syto.digital/
Also, bulding a marketing tool to help local businesses be ranked by ChatGPT: https://www.pluspoint.io/
Still working on my Minecraft MCP project. The models have good image analysis now so I'm feeding them topographic maps of the terrain and they build using Starlark scripts.
This keeps the tool calls low and it's fast enough to work in voice mode.
https://www.youtube.com/watch?v=FxZU1DuSEGg
I've been working on a port of Pokemon Crystal for a specific ESP32 dev board with a round amoled screen.
The goal was to have a basic pokemon themed tamagotchi clone with step counting that motivated me to walk more. It snowballed into a full fledged rust port of the whole game, with a completely redesigned UI meant for round screens and touch controls. It's got bluetooth and wifi trading, autosave, contextual action buttons, etc.
The most experimental and interesting feature is an auto play mode, which keeps the game advancing while you walk in real life, letting you take over at any time and keeping a log of any interesting event along the way.
Not launched yet, but I have a very preliminar landing page for it here:
https://pokehike.bernalr.com/
I'm building an open-source "spatial workspace for directing AI coding agents like an engineering director".
I got overwhelmed by walls of text of parallel Claude Code sessions. This limited the sessions I could steer in parallel to about 5. I knew there had to be a better way, and I found one: A canvas per project. Work laid out as cards. An agent in a worktree per card. Chat condensed to WhatsApp-style back and forth, not a litany of tool calls. Result is usually a narrated demo video that I can comment on by voice to have the agent go another round. Remote-control from my phone. Remote workers to scale beyond my machine. Everyone I know who tried it has stuck with it so far.
https://github.com/sadilek/obeya – looking for feedback!
A vibe-coded app that shows all your Android Health Connect data in one place. I use it daily because I have a garmin watch, withings scale, samsung health for nutrition data, liftosaur for workouts etc.
Started it mainly as a way to use AI and I named it after my dog to immortalise him: https://play.google.com/store/apps/details?id=com.gads.hamil...
I also priced it cheap with no recurring fees.
I'm building a self-hosted media server for video, like Plex or Jellyfin. I'll be opening the beta in about a month.
Features:
- No paywalls, bandwidth limiting, or crippled behavior. No login required.
- Single go binary, deployed via Docker.
- Reaches your home server from anywhere with no port forwarding, no VPN and no paywall, even behind carrier-grade NAT. No Tailscale required.
- Cast to another device. You can use any logged in device as a playback screen.
- Watch together, with voice support: Play synced video with friends, and also chat voice. You can also mix this with casting to another device.
- Dialogue boost and loudness levelling for every film.
- Real adaptive quality, tuned for a home's upload.
- Hardware transcoding, full quality in every supported browser.
- A library that explains itself: what the scanner saw, why each title matched, guesses marked for you to check, and a disk that's gone missing shown as missing instead of its titles silently vanishing.
- Your data stays yours: The database is a rebuildable index, metadata corrections can be written back as NFO files that Kodi and Jellyfin read, and an automatic backup keeps what can't be re-derived (watch state, preferences, shares).
Security is a primary concern:
With the ever increasing pace of hacks and CVEs and AI-assisted intrusion, you simply can't be too careful anymore.
- The core (which talks to the network, handles the API, users and the database) never opens a media file or runs FFmpeg.
- A separate media node does every file read and every transcode. It has no network access and talks to the core only over Unix sockets.
- FFmpeg runs under Landlock and seccomp with no path to the file and no socket. All it can do is read and write streams.
- Small dependency tree on an allowlist, pure Go with no cgo, secrets sealed at rest and never logged.
- The library can be mounted read-only, so even a full compromise can't encrypt or delete your media.
- Cryptographically signed chain for servers and releases.
https://orpheum-app.com/
Love the Go selection for this ( I am a recent convert myself ). Be sure to plop it on show hn when it is ready.
Would love to check it out. What will your monetization plan be?
I was chatting with a user of Eternal Vault (eternalvault.app), and we discussed about a grave problem that I always knew we had, but I just wasn't 100% sure how it could be solved.
It was about what happens if Eternal Vault itself shuts down where people are putting their trust in something to be there when the time comes when they are no more and their trusted contacts wants to unlock their vault.
We chatted on reddit, and I majorly worked towards making it all work even if Eternal Vault went under someday (I hope it doesn't)
This is what I am thinking about: https://eternalvault.app/continuity/
Lipwise (https://getlipwise.com) is a narration app with an LLM in the pipeline capable of fixing mistakes and doing transformations, and the best thing is even with a mediocre machine it can be fully local. No telemetry, lightweight and fully open source https://github.com/lnenad/lipwise
Rebuilding my Covid-era realtime social gaming/quiz platform. It’s been steadily growing over past few years (380k people total).
It’s always been a hobby to get people socialising/connecting more, and I’ve been wanting to build something that can put Convex (almost like Firebase) to good use since discovering it.
https://ziago.co
My own programming language called: dia (stands for do it all).
Aims to be a full batteries included language + standard libraries, so that building both with and without AI assistance is effortless and can reuse the building blocks without having to reimplement things. Wrote more about the why here: https://dia-lang.org/vision
Most notable such case is the UI library, where you can take out just text rendering, or event/input system, or the signals-based reactive library, and reuse them in other contexts.
some other notable features: - vey inspired by rust in syntax and type features - batteries included library containing a full cross-platform gui library including a prosemirror-like wydiwyg text editor - rust like memory safety, uses arenas for cyclic objects so that the language doesn't need lifetime parameters like rust, easier to reason about for generic code - transpiles down to c, then compiles to anywhere C can run - has its own native JSX-like syntax called DSX, any functions with the right signature can be written in DSX notation, not just UI components - incrementally compiled, has HMR (hot module reload) - effortless C interop (dia compiler ships with clang). no need to write manual wrappers for libraries (but sometimes need to write a short manifest file to map dia's memory safety with most c libraries' manual management)
More information here: https://dia-lang.org/ (Not open-sourced yet, waiting until the base features are stable)
I have finally released my quirky wishlist app, https://thingstohave.app, that I was covering in multiple previous "Ask HN" posts.
Right now, I'm busy with marketing, so there's not much code stuff I can tell you. But here are some marketing finds:
1. At this point, ChatGPT traffic is competing with paid traffic, while having a much better audience quality and retention.
2. I had a disastrous Product Hunt launch, despite having two badges on previous launches. It seems like the platform went into "dead internet theory" territory. I wrote a rant about it here: https://www.brachkow.com/notes/you-should-put-less-effort-in...
And as for code, right now I'm busy with hobby projects, and two of them are:
1. v4 of my Inertia.js adapter for the Hono framework – https://github.com/brachkow/hono-inertia/pull/6
I want to expand the thing I made to write Things To Have without relying on annoying "full-stack" frameworks, into:
- the best Inertia adapter for Hono
- the best way to do SSR in Vue (or whatever framework you use)
If you don't know what Inertia is, check it out, as it may change your frontend life – https://inertiajs.com/, and Hono is modern express.js – https://hono.dev/
2. I'm building a mobile multiplexer because every one I tried was either too old-fashioned or low quality.
Now I've built a great app with good scrolling, good composing, simple use of hotkeys, file sharing, and many, many QoL features. I'm really using it on the go almost daily across two months now.
The only thing that slightly set me back is that I went too deep into supporting the herdr-tui protocol, and herdr moved into a new protocol that I'm not fond of.
Things to Have looks pretty nice; congrats on launching!
It's missing one feature that means I can't migrate from GiftList: allowing multiple accounts to edit a single list.
In addition to my own list, I'm managing lists for my kids. Their mom also needs to be able to manage them. On GiftList she and I can each have our own accounts, I can create lists for our kids, and then I can invite her as a collaborator. I didn't see a way to do something similar on Things to Have.
Also, for purposes of your comparison table- GiftList has an export to csv option.
Thanks! This feature is in making – if you reach me out on support email (can't leave it here, as it will unleash a wave of spam – you can find it in the end of https://thingstohave.app/help), I will be able to open it to you under flag for testing at the end of this week.
Regarding, comparison – I will update it. I have some fact-checking planned soon, as more apps started to offer MCP, and these SEO pages became quite dusty:)
I built an app to help me develop healthy habits and routines on my computer. IT blocks sites/apps based on a schedule and tracks my activity for later analysis.
I had simpler versions of this earlier but never something comprehensive. It was also my first AI assisted development project. I've been using it personally for around a year. I polished it and released it as an open source project last month. It's specific to Linux (which is my primary platform)
The website is at https://glockerapp.com/
I'd posted this earlier on HN hoping to get some feedback but didn't get any traction https://news.ycombinator.com/item?id=50349646
I quit my job a year ago to launch https://docuview.ai.
It’s been sort of a whiplash experience, but ive been so much happier building this business than I was working.
Real estate is a space that’s very close to me, so I was able to see the demand for a product like this long before I went searching for it.
I’m not alone in the space, but we’ve focused on building a product that’s opinionated enough to where the other players would be taking a risk by trying to copy the things we do. I think this was a good move.
We won’t be the largest player, but we don’t have to be. And we build the product we ourselves want, not a product that kinda fits everyone but every user only uses 20% of.
I’ll probably never go back to a job :)
Porting marginalia's index code to C++. Been staring at the code for so long I've found optimizations that will drastically cut pointer chasing, necessitating a primarily io_uring-based buffer pool population strategy with rare naked preads.
Almost 12 years on Micro https://micro.mu. something I keep trying to kill and won't go away. Started as a Go framework now more of a personal assistant and home server. I stepped away for a few years and came back to it to try fulfil the goals of the original mission which was a Google alternative or superapp. It's not really for anyone else at the moment even though I post it here frequently. Probably needs a ground up rewrite. Currently working on an android app. It also has a pwa.
What are you working on? https://wpmaven.ai Describe your business and WPMaven.ai creates it, runs it and grows it for you. It does all your marketing, including UGC videos on all social media, sales follow-up and closes sales for you.
You can even manage your business on your WhatsApp! You receive your weekly report in your phone directly and you approve or edit anything that goes online (because yes, AI is not perfect yet).
We launched recently and are polishing everything while finding the optimal verticals (if you guys know WHERE would you use it, let me know!).
I’ve been working on Beacon, which just got approved on the App Store! It’s a mobile app to instantly answer “which friend is free to talk right now?” and summon them in a single tap.
Send your beacon to a group, and everyone gets a gentle ping. The first person who answers gets connected for a 1-1 call, and for everyone else the signal drops silently. It does not log any missed calls so there’s neither pressure to answer nor pressure to return calls.
Overall it works really well! It’s best if you're able to join with 4+ friends/family you’d like to chat with more. I have dozens of connections in the app now and it feels like absolute magic to me. I tap the button to send my beacon to everyone and get a random friend on the line.
https://trybeacon.chat/
This is really cool! I'm very excited to try it - nice to see it available on Android also. As someone who has messed around with building relationship-focused social apps, I'm curious if there were any other concepts you tried, or other apps you like? What did you think of BeReal for example?
This looks amazing, and your design is really cool! Congratulations, I hope it’s successful however you define it!
Hey HN!
My co-founder Abhishek and I are master’s students at the University of Ottawa who got tired of buying cheap furniture every time we moved apartments.
We decided to build Usefulls ..starting with the Nomad 23, a standard 23 kg checked suitcase that physically converts into a twin-door wardrobe cabinet. When you take the removable wheels off, circular bumps on the base seat into grooves on top of another case, letting you stack them into a full-size modular wardrobe.
We’d love your feedback on the concept, the physical design, or how we're approaching modular furniture/luggage!
You can check it out here: https://usefulls.com/
This is a sick idea!
I am co founder Abhishek , Thanks for replying We would love if you can give feedback on our website usefulls.com
A while ago the invoicing app I used (Harvest) jacked up their price about 10x, so I decided to build one for myself.
And I did: https://hourbun.com/
Just an intuitive and fast time tracking and invoicing app built for people and teams like me who want a quick and reliable way of invoicing clients without paying a premium or suffer through bad UI/UX.
I'm working on Hale, an abstract distributed systems language.
It's a traditional systems language in the sense that you have fine control over how resources are used, but it's abstract in the sense that you have to come to an agreement with the compiler about your intent around the resources, rather than directly declaring the machinery.
It's a distributed language in the sense that there is a higher level structural abstraction in the language that situates computations of the program within a recursive graph structure. Part of that structure incorporates message broker semantics. This gives the compiler a unified view of programs across arbitrary process boundaries. The structural abstractions also enable an effects+formal verification syntax within the language, which allows programmers to codify architectural decisions that are then checked by the compiler.
The compiler itself is infrastructure aware, and the highest layer of functionality it handles is configuring the environment into which applications deploy. The single language handles everything from individual memory allocations to distributed fleet deployments.
The language is close to v1, and I'm currently building a fully distributed inference-as-a-service application with it. Arbitrary hardware/supply matched against arbitrary callers/demand in real time.
https://github.com/hale-lang/hale https://github.com/hale-lang/voice
Still working on Kavla, my open source SQL canvas https://github.com/aleda145/kavla
I'm very excited by adding interactivity. I have a new "knob" feature that pops out when you type {var}. It infers the column's type and values.
`SELECT * FROM hn_posts WHERE score>{score}` will grant you a slider that goes from 0 to 6015
Titanic Browser demo (no login, it's all WASM): https://app.kavla.dev/
Also been working on a "chat to data ingestion pipeline". Not sure where I'm going with this but I like the idea? https://github.com/aleda145/pompos
My god, as a Greek, this name is hilariously lewd.
Haha so I've heard! Mistake on my part, but I kinda dig it
For the record, it means "to roll something out to a flat surface" in Swedish
My pet project is this match tracker for League of Legends: https://scout-for-lol.com
Essentially it posts notifications to Discord when someone starts/ends a game with a pretty image.
All your games are saved to a datalake and you can create competitions or scheduled reports over your community's matches.
I've been working on this for ~4-5 years.
I'm working on a diffusion inference engine. I dislike node editors like ComfyUI and wanted a decent python API to control it. The first version was another python/PyTorch variant which was decent but still way too slow to my liking.
I ended up rewriting all the inference in C++/CUDA and I'm finally getting close to limits on the devices I tested. The benchmarks are still a work in progress, I plan to add high resolution comparisons (1536^2 for images, and 1344x768 for video) to give insight in how well this scales.
https://noxengine.dev/
I've been disliking the recent trend of LLMs releasing with very promising and high benchmarks, only to disappoint by failing simple tasks or sanity checks. Plus, benchmarks don't really tell you about the LLMs' stylistic choices.
So I made 'Model Style not-a-Bench' to be able to quickly eyeball various LLMs, most preferably at 4-bit quants, under 100GB weights (excluding storage offloads like Qwen Flash Next's approach).
https://msnab.pages.dev
ArtCraft and the Craft Apps. https://getartcraft.com
We're going to open source everything. Software, SaaS, infrastructure, weights, platforms, protocols.
We're going to bridge local and cloud compute in a seamless way.
We're building an app factory for high throughout development. Reverse engineering tools for non-technical people. We're going to make high quality libraries that others can remix.
We're going to make all of this agentic and AI/CLI first class.
We're going to open source everything and make the world hacker-first again, like Web 1.0 with P2P and RSS and APIs and IRC.
Join us!
https://discord.gg/artcraft
Im in awe of you folks' progress. All the best.
I’m working on ai interviews for system design prep.
I’m currently in a beta phase giving people free access for feedback. If you’re prepping for interviews and want to give it a try let me know
https://liveround.ai/
It was born from the fact that whenever I prep for system design I spend a lot of time revising and learning material (as well as drawing from my experience) but whenever it came to the interview I just seemed to struggle to put that together to design the system I was given.
For Halloween I made a VR survival horror game for meta quest 3/3s.
- https://sxp.studio/aftercorp
The cool thing about that game is that it uses a realtime environment mesh reconstruction algorithm that allows me to create an approximation of the world 3D geometry and project a dummy “flashlight” with hand tracking. I think I’ve only seen one other game do that technique on quest (some laser tag game).
It runs at 60fps with a tight GPU budget. Consequently it’s not the most beautiful game, but it’s actually pretty fun, immersive and can scale to all sorts of environments. I’ll be porting it to iOS next (handheld AR style).
I'm making a replacement motherboard for the Nokia 1100 that runs Linux and supports LTE. I have the software working well on a MangoPi dev board and I've designed two PCBs: one "test fit" board that just breaks out the phone keypad/LCD to pin headers, and the actual board with everything on it. Now I just need to order the PCBA which is super expensive :)
It uses an allwinner T113-S3 SoC and the Quectel EC25 LTE module, and I've made sure voice calling works over LTE, at least on T-Mobile in the US. (this is tricky bc of carrier rules)
video and photos of MangoPi test setup and my board layout here: https://constcast.org/n1100c/
Interesting.. May I ask why pick such a simple model of Nokia if you'll tinker with it? Just to learn?
I wanted the challenge of trying to fit all the components into the small board area available (this phone is tiny), and I liked the idea of creating an updated version of the best-selling cell phone of all time. I figured some of the "dumbphone" crowd might be into it too, if my design works I might make them available for sale
I've built two directory app websites as a fun way to track some topics that I’ve been interest in.
One is https://bannedbookshelf.com which lists books that have been banned for whatever reason, or have been yanked off Amazon for being too ‘spicy’.
The other one is https://fangamesdb.com which tracks the decompilation and recompilation scene which is exploding now with AI reverse engineering tools.
For the past year and half or so, I have been building https://rationalgrid.ai with the goal of bringing together a multi-player LLM interface (based on a underlying graph data structure) and publishing platform, for people to work together to explore deep questions.
It has been tough going, and not really found an audience yet, but would appreciate any thoughts / criticism's.
(Built using Elixir / Phoenix / Live-view / Cytoscape - an amazing stack for web / AI)
WhatsApp released the Agent API but it needs a constant poller to run which sucks as all my apps are serverless that scale down to zero when there's no traffic. So I built Unpoll (unpoll.cc) that takes care of polling, cursor, webhooks and retries. You can simply add your agent's token which is stored encrypted and then send & receive messages by plain old HTTP requests and webhooks.
Lemme know what you guys think and happy to provide a free month to you as beta testers!
Two of my flagship projects I've been working on lately:
Video Commander - A desktop application suite of video encoding, inspection and quality analysis tools (ffmpeg, vmaf, mp4box). I now have quite a bit of paying customers!
https://video-commander.com
Cargocat - Also a desktop S3 client that I created out of some of the components I've done on Video Commander. Newly launched, so still slowly gaining some traction.
https://cargocat.app
Both are built with Rust + Tauri and React. Happy to answer any questions.
Working on / dog fooding my graph based recipe app[0].
There are more things I'd like to do with the graph concept, like linking two user's recipe trees if they were imported via the same URL. Really trying to to avoid features that I think would be cool that don't provide any functionality.
There are still a few features I'd like to implement that are more in the vein of nice to have in any recipe application, like nested recipes, temperature conversions, volumetric ingredients amounts to mass conversions, etc.
But I've been using it for new recipes and converting my old recipes I have written down in the margins to this format and I've been happy with how usable its been in these early stages.
[0] https://grainofsalt.app
A context cost driven software design theory https://www.contextcost.dev/cmp/
A small tema and I are currently building https://scenema.ai, a platform for creating whiteboard and 2D animated explainers. We initially started Scenema to deal with character and voice consistency in cinematic AI videos. Most video models can only generate 5-10 second clips. More recently that has gone up to 15s. If you wanted to create any meaningful video longer than that, it becomes incredibly difficult to manage your characters and voices.
We ended up moving away from cinematic films and started focusing on explainers for two reasons: 1) we consume a lot of explainers and wanted to produce our own, and 2) traditional animated explainers are very expensive to produce and time consuming.
It took us a few months to get here but we’re now able to generate minutes long explainers ( up to 30 minutes) with studio-level animation across whiteboard and 2d. Here are some examples:
https://scenema.ai/explore/what-is-audio-compression-dinner-...
https://scenema.ai/explore/do-opposites-really-attract-what-...
https://scenema.ai/explore/every-salon-service-is-a-chemistr...
https://scenema.ai/explore
I got tired of always having to update my wife, and my family members résumés. So I created https://theresumechef.com It's private,free to use, no pay walls or monthly subscriptions.It has10 resume templates to choose from to get you started.The final resume can be exported as a PDF, or stored on Google Drive or Dropbox in JSON format. There are still a few issues, but I've learned a ton by just creating and launching this project. I have some exciting new product ideas, but I'm currently waiting on my Dutch KvK number to accept online payments with Stripe.
I'm working on an IoT automation platform built on top of AtomVM that lets AI agents control a fleet of devices, including device management, telemetry, OTA updates, etc.
The idea is to expose device capabilities (sensors, actuators, cameras) as tools via MCP, effectively giving AI agents the ability to interact with the physical world.
Currently, the platform supports all flavours of the ESP32 and has partial support for the RPi Pico 2 W.
The backend is Elixir/Phoenix, the firmware is Gleam with occasional Erlang.
I'm working on a shared calendar for couples. I attempt to build a distraction-free experience inside the app and to keep it as a PWA.
https://joinsundown.app/
Django + HTMX + SQLite, Hetzner VPS - feedback is welcome.
I’m building a voice-first second brain for Apple Watch. The idea is simple: capture thoughts instantly from your wrist using the Ultra’s Action button or Siri keywords, without needing your iPhone nearby (using cellular or wi-fi).
Audio is transcribed using EU-based speech-to-text with zero data retention, then turned into searchable notes and summaries.
Basically, I want to capture a thought before I forget it, without pulling out my phone.
If anyone’s interested, I might consider publishing on the app store.
In order to capture instantly, how do you open the app fast? Does the Apple Watch have a wakeword you can utter for that?
For now, the action button on my watch ultra triggers a shortcut the app offers. The shortcut launches the app and starts recording immediately. Takes less than a second. When the action button is pushed while recording, it stops the recording. Quite smooth workflow.
I’m trying to make Siri to trigger the same shortcut with just a voice command but somehow Siri doesn’t do it so that part is still work in progress
I've been working on a pi extension for running agent swarms. Paired with deepseek, this has been performing really well for various research tasks.
https://pi.dev/packages/agent-horde
After every run, it generates a postmortem which I then use to iterate on the protocol. I call it poor man's RSI.
Hmm. A lot of smaller projects that seem to be slowly integrated into agent skills. I am admittedly moving into agent world very slowly.
Though now, with the specter of raised prices on the horizon, the local only portion of my setup gained some additional focus.
Still, from the fun mini-projects:
- attempting to replicate testing whether agents elevate inferred persona, memory, or background context above explicit user requests ( https://bloomberg.com/news/newsletters/2026-10-07/study-clau... ) - a few hackaton entries - my llm rpg got some few additions ( still nothing I am willing to publish ) - first working and useful unraid app;p
I'm currently building AssetJay, an asset management tools for companies without IT departments. When I grew my startup to 25 people, I always tried and failed to keep track of all the devices and peripherals we owned, so built https://assetjay.com to make it easy to do.
https://evalship.com is what I'm spending my time on most - approaching it as Trello for evals; a platform that should be good enough for most people.
Also spending too much time on https://extraheadroom.com to support existing users
and too little time on passion projects https://nommer.ai and https://criticaster.com
It's been a productive year :-)
A site blocker extension for Chrome and Firefox. Fast and not recording any data. I was sad that I found an easy to use one or one that has no recorded my usage: https://github.com/SiGe11/Lightweight-Blocker-Extension
This one I (mostly) did without LLMs, but the other project is vibe coded, but I'm enjoying it. An RTS made by Claude in Unity: https://github.com/SiGe11/StarForge2
Whoa, StarForge2 looks amazing. I cannot believe (well, I can, lol) AI has created a game as good as or better than commercial RTS from major studios.
https://sloppy.live
I gave Claude Code a live stream and a voice. It already build 1700+ games and apps live based on suggestions from chat. It also controls its own appearance and the website itself.
My favorite creation is Heidelberg Blocks. What started as a 3d bridge is now a 3d world with multiple German cities, a karting track, a paragliding hill, a flight simulator and much more.
Really fun side project.
https://inversion-works.com/
An app-only lunch bowl chain that can execute very complex customisation without breaking. We've built "cook augmentation" that takes over decision making and QA, added an app-only ordering (accessible to the agents) and ended up with a frictionless experience where you just tell your agent "I'm hungry" and we cook a fresh, hot, delicious bowl that's good for you.
Currently at prototype stage, raising angel round in order to do a live pop-up.
I am rewriting LaTeX from scratch in zig. Byte compatible with pdfTeX XeteX and luaTeX.
It’s faster at compiling, better errors with modern DX.
http://www.voltlatex.com
My custom harness Xi - complete with code mode in clojure, guarding by rules, mult player etc: https://xi-agent.com/
I am building a work/coding agent with my consulting client(s) in mind. They are a non profit and have a software (that I build) to manage property tax exemption compliance.
They have 70+ property managers as clients in the states of NC/SC. A few of the property managers are quite large, managing 10000+ units. The property managers use a mix of Entrata, Yardi, custom Excel to manage the source data. Only a part of that data is needed for tax compliance (which is what I build).
I provide help onboard the property managers. I do all my work remotely, living in India. Lately, my client (the non profit) has been creating mocks of new feature ideas and sending them to me. They use Claude desktop I think. They also maintain a Google Sheet with bugs or new feature requests. I manage the entire development and support via Claude Code.
My idea is to give my client a desktop app to directly manage certain aspects of the software. For example, we support many different Excel/CSV formats for the incoming data. These are exports from Entrata, Yardi or manually managed Excel. Our expected format is fixed but on many occasions there is data missing or poorly managed upstream. My client can easily figure out what is wrong if they get access to the files and generate tasks for coding agent which I can review. Or, if there is data missing, they can also send emails to the property managers. All property managers (my client's clients) could also be given a desktop/web app to help managing the data at source.
I would host a backend for the work/coding agent on a server. The desktop/web app would manage users, permissions and tasks between users and agents. Hopefully, this would grow into a much larger AI enabled product for property managers in a couple years.
Lipik, a free open-source dictation app for Mac (https://github.com/Gourav2609/lipik). Hold fn, talk, let go, and clean text lands in any app: ums and self-corrections removed, spoken punctuation turned into real punctuation. Speech-to-text and clean-up both run on the Mac. On my 106 English test clips, lines that came out fully right went from 28% to 76% with clean-up on. Right now I'm working on Hinglish (Hindi + English in Roman letters). It gets 7 of 20 test lines fully right, so there's a long way to go.
https://independentvet.com
Independent Vet - a website and app that helps pet parents find locally owned (and corporate/PE owned) veterinary practices nearby. It came out of a bad experience we had with our dog 3 years ago at a practice that had been recently acquired by private equity. The sites now get about 30k visitors a day and continuing to build in the evenings and weekends with Claude/Codex has been a lot of fun.
As many of you, I have a few subscriptions to Claude and Codex (just me??). It's a daily occurrence to use 2-3 accounts for the same conversation (whenever the limit stops me).
What I need is something to switch the account where the usage is taken from, not the configuration.
I made something simple & transparent named Switcheroo: https://github.com/FTCHD/switcheroo
Works with other AI CLIs like Gemini, and other various CLIs like Vercel (if you're managing multiple accounts like me), Netlify, Github etc
Surely against terms of service, no?
I'm not sure providers would shoot themselves in the foot by not allowing you to buy more usage. It seems like both OAI and Anthropic could use all the subs they can get.
If that's not right, blame it on Chat:
"Yes, OpenAI explicitly supports having multiple paid ChatGPT accounts. Each account has its own subscription, conversations, and usage allowance."
I continue to work on Akariq my travel eSIM business [0]. No app or account required. I have expanded my offering to 200+ countries. In the next 1 week to so prices will come down further in Europe and Caribbean regions.
A lot of people from HN have bought travel eSIMs. I’ve received messages of support plus most people have switched over from Airalo which is expensive and generally doesn’t connect to the best network in the country.
[0] https://akariq.com/en/
A 3D pixel-art basketball game built with JavaScript
Play: https://vladsez.github.io/iso-hoops/
Source: https://github.com/VladSez/iso-hoops
Features:
- 5 game modes (free shooting, 10 free throws, HORSE, beat the clock, 3pt contest)
- 5 courts with dynamic day/night cycles
- offline
- no backend
Camera Nocta. I started making an Android app to take better black and white photos of my wife, and it snowballed into a full fledged film-styles app for both Android and iPhone (if I ever get app store approval). It's best for portraits (tuned for a lot of skin) and fashion photography.
The RAW pipeline is my own, with my own denoise, because I hate the "skin waxing" that bad noise reduction and AI seem to do. So there's lots of natural grain. The UI is somewhat opinionated because geared towards more shooting and less editing. The editing tools are my own ideas regarding how to push color (more like dye and pigment than like saturation tools).
It's out on Android, but the Test Flight has more users (1,478) than the released Android app I think. Every single artist that showed interest online used an iPhone, so I guess I should have expected that.
https://play.google.com/store/apps/details?id=com.cafesarris...
https://testflight.apple.com/join/Jtppv1Xc
Personal project - long past due to set up a home lab so I'm piecing together a SAN and some compute from old parts i have laying around and some second hand servers in a new rack I put together, along with setting up a bespoke home network instead of just using ISP stuff.
I have long been averse to configuration and customization but my needs and preferences have outgrown commodity systems and default setups, and AI is making it much easier to customize systems, so I'm embarking on this journey after over 35 years of using computers.
A friend needed to decide which cybersecurity conferences are worth going to, so I started building Find a Conference (https://app.findaconference.net).
I'm more of a UX designer / frontend dev, and I've had to learn a lot about things like postgres queues, auth and CI/CD in the process of making it. It's been a challenge trying to develop the features, while also maintaining the conference list and verifying all the info (I use some AI in the data sourcing, but want to keep it all human validated). I'd say it's just reached MVP, and it's definitely still a bit rough around the edges, but I'm already seeing some organic traffic. I'm also learning some conferences are very expensive...
Open to any feedback, especially from folks in the infosec industry!
https://killswitch-lang.org - a programming language designed to be impossible for LLMs to read/edit/write
https://en.wikipedia.org/wiki/Malbolge
I’m seeing if an agent can run a full e-commerce website. Nothing fancy, just a storefront that connects to a print on demand supplier. The agent releases new products and designs and makes sure there’s no hiccups in the orders.
https://agent-commerce-teal.vercel.app/
I am working on a casual game / experience about Indian festivals as a hobby project. This started as a Diwali night. You can walk through an old neighbourhood down to a river, light 108 oil lamps, set of crackers and send lanterns to a sky that is shared by everyone. It runs in browser without login or any ads.
I have recently added more features like a lamp you keep burning (as per Indian tradition) & a rooftop garden that keeps growoing. I have ideas to add more missions - planning to continue building on this over the year.
Some stats: Around 1,100 players have lit all 108 oil lamps (diyas) & 1600 lanterns released into the sky. Yesterday was the busiest day (because it was Diwali festival)
URL: https://virtualdiwali.com
I am building https://withthislink.com
It's a site for couples to create a wedding invitation website where guests can RSVP right on the invitation. You get a personalized link for an invitation with different domain options.
I think it looks neat and turned out rather nice.
I’m refactoring https://rcarmo.github.io/projects/piclaw around pi-durable (which is essentially removing overhead and duplicate code at this point, since it was already conceptually close) and wrapping Obsidian into a server-side app with a pi-durable agent to help tag and organize notes. I also have a few Go ports of things like pi-ai and core SIMD inference stuff going, slowly, since it’s too easy to get sidetracked.
I've been laid off in December, and this year I've been trying to get developers off Gitkraken and Termius subscriptions by building 2 Rust native interconnected apps.
You can check them out at: https://branchbay.dev
I've been working on trying to be more helpful around the house, and trying to spend more time with my kids. I'm trying to give my partner a little bit of relief so that they can pursue their passions as well.
so wholesome <3
I've been using the recent MIT weight-space programming techniques to work with neural quantum states, which are neural networks that represent a quantum wavefunction. I was able to produce a bunch of novel networks for a few hundred dollars of cloud GPU, which would have probably cost tens of thousands to train in the traditional manner.
https://paulg.info/2026/11/04/quantum-weight-space/
I'm really pleased with how my equivalent to Smart Objects turned out in Mojave Studio, my Mac image editor. In Photoshop you can transform an opaque sub-document with an affine and/or homographic transform, while in Mojave Studio you get what I call a "Smart Folder" shown inline, as if you're examining shadow DOM in web inspector. The thing you're viewing has its identity transform and the rest of the document gets the inverse transform applied, to see how it sits in context. https://mojavepaint.app/blog/smart-folders
Late the bandwagon but have been experimenting with agent assisted lean/formalization for the first time; it's been surprisingly high ROI vs previous fuzz coverage and has forced me to think through some interactions of cases that I hadn't yet encountered in the wild. Very useful for flushing out null behavior in my typed SQL engine [1].
On another project to inventory street trees - making progress on training a vision model for aerial tree photo detection in urban environments (original goal was to fill in central park in NYC, which is missing from their public inventory). Haven't gotten NYC to the level of precision I want, but SF has good initial results! [2]
[1] https://trilogydata.dev/
[2] https://arborary.world/cities/ussfo.html
final polishing of FinditAgent, a Windows tool which allows AI agents to research large collections of local files without vector database (more like grep on steroids, for almost any file-format) - and without giving the AI any access to local file systems.
The user may ask any question, the AI will create the search strategy and queries, FinditAgent will act as a kind of Archivar, doing the searches and sending only results or utf8-excerpts to the AI. The tool offers predefined connections to the api of most cloud models - or to local running AI via ollama.
I have developed the underlying search engine for more than 30 years, now added the agent integration. It's now available on GitHub, free for private and experimental use. At this point I'm mainly polishing the documentation and testing how reliably it works on archives other than my own.
https://github.com/MHoub/Findit-Agent-Frontend
Its a one click windows install and I am very interested in hearing from people who work with large, messy document collections.
https://lapyn.com/
It's a product design platform where humans and AI agents can work on the same project.
The main idea is to have a platform-independent system of record for your product that agents can interact with and use to generate code for any platform.
With agents generating more and more code, I think it's important to have a place where the actual product decisions live. Not just the design, but also accessibility, translations, behavior and so on.
I’m building a personal apps platform! I still am hoping for a future where every interface doesn’t start with a chatbot.
https://github.com/momja/exhibit
You can use an agent to build small apps for you, your friends, family, etc. and use them on the platform. You can also download them and take it with you so there’s no lock-in or special sauce. Its deny-by-default security policy helps you wrangle untrusted AI code and I’m iterating on API connectors to bring your data in. This week specifically, I’ve been working on adding some UX polish to it, figuring out how I can run a reliable hosted version, but you can self host it today. I’m not a designer, so trying to learn a lot and leverage any tools I can!
I've always been interested in how "AI Assistants" work. There's now a ton of them out there. But you can run mine locally, and it has a couple of easter eggs ;-) as well as voice notes through telegram, connect MCP servers, and runs in a single Docker container needing less than 100MB of memory. Any issues/feature requests would be welcome:
https://github.com/mutablelogic/go-llm
No access to email/calendar directly, so that's the next thing.
Just like the last time I posted in this thread, I’m working on The Marvelous Brass Chessplaying Automaton, a chess database/playing/opening training app for Apple platforms:
https://testflight.apple.com/join/mqfhpmHY
The original goal was an alternative to ChessBase, but with an Apple-native GUI and iCloud sync between devices, and as an improved alternative to tools like chessbook.com for operating repertoire management and spaced repetition opening memorisation. In recent months I have focused more on the chess-playing parts of the app, trying to implement human-like computer opponents at all levels, using a transformer-based model (similar to the one described in the Maia 3 paper) combined with a very small search.
Recently I’ve also been working on chess variants. Since a while, the companion Apple Watch app plays 5x5 chess. More recently, I added a 10x10 chess as an Easter egg game inside the app, and I’m now working on hexagonal chess. With the current state of the art in computer chess, it’s almost trivial to implement vastly superhuman strength in all these variants, but implementing weaker levels that are enjoyable opponents is interesting and challenging. Unlike in normal chess, there are no big databases of human games to train on.
I just launched a no-subscription fitness tracking app (macros, biometrics, workouts, supplements, etc.), aimed at bodybuilders. https://apps.apple.com/us/app/cheiron/id6825576174
I'm working on the interview portion of my federal income tax formalization project. I've hit my coverage target (80% of filers), so I'm doing some refactoring to optimize future development. https://github.com/emmilco/us-tax-model
Privately I've been running a "game" for the past couple months with claude where it asks me questions about movies I've seen, tries to predict my answers, I grade the answers, and then it builds a theory of my movie taste. It's gotten really good and has reached the point where it can reliably recommend movies and shows I actually enjoy.
RainbowSay https://rainbowsay.ai/
I used to watch K-drama on Netflix and pause it to copy paste subtitles in AI chat to do vocabulary breakdown, translation and grammar. So I automated it in an app which auto pauses video when it detects a sentence and gives translation, vocabulary (for every word) and grammar.
I also added same to voice from microphone, for deliberate practice speaking what was learned. To help with spaced repetition.
For about 10 months or so I've been working on this FOSS hobby project to get more practice at Linux packaging, cargo utilities, and migrating to Nix and NixOS (as well as solve some personal pain points in using a HDMI TVs and projectors with PCs).
https://tisvlay.readthedocs.io/
> TisVlay brings TV Display control for Linux, including: power, volume, backlight, AV mute, input switching plus more, with per-device profile automation via desktop session integration (idle/lock, boot/suspend, remote login/streaming, etc), and in reverse, letting the TV lock, sleep or wake the PC and its remote drive the desktop, while supporting various vendor interface protocols for multi-display setups, mitigating energy usage and OLED panel burn-in or projector bulb burn-out.
Disclosure: this project includes AI assisted coding, documentation and research, and has been a great help given physical disabilities in typing. After programming by hand well over a decade, using voice dictation scene the '90s (IBM ViaVoice or Dragon NaturallySpeaking anyone?), and worsening hands over after my CSE doctorate, this was also a bit of an individual research experiment to explore how effective LLMs could augment my personal approach to spec & test driven development.
While still beta (pre v1.0), its already daily drivable and fully documented. Being largely feature-complete, my current focus has shifted towards distro packaging, and custom ATmega firmware in Rust for HDMI CEC USB adapters to solve a few TV firmware bugs. If you'd like to give it a try, the installation guide and docs are linked below. Or if not, and you already use something else, I'd like to give a shout-out to netflector, especially for privacy-conscious home-automation users, given how common spyware is in smart TVs and projectors these days. The docs cover using it to keep displays on an isolated network.
https://github.com/netflector/netflector
https://tisvlay.readthedocs.io/en/latest/operations/isolated...
I am trying to build a better place for people and companies in tech to find jobs and talent. Recently I redesigned the whole platform but everything is pretty much work in progress.
https://thedream.work/
https://uruky.com is a private, paid, ad-free and personally customizable search engine. It was launched by my wife and I earlier this year and reached 400 MAUs and €2000 MRR last month.
https://podby.com is a simple EU static and app hosting service. It was launched last week, and is at €20 MRR.
https://fairats.com is a simple and fair Applicant Tracking System for remote-first teams, launched a couple of months ago, still at €0 MRR.
Happy to answer any questions and consider any suggestions!
Any benefit / difference over Kagi? Congrats on the growth.
Thank you! It depends on what you care about, because it has less features (it's not trying to be an ecosystem with Mail, Browser, etc., and it doesn't have AI features, for example), but you can control which search providers to use and you get access to the source code after paying for 12 months, which Kagi doesn't do.
Working on Aqueous [1], a next-generation compressed archive format. I've just finished a long refactoring session that enables streamed processing instead of keeping intermediate results into the memory or spill file, and am now working on formats that were big enough that required this support (e.g. disk images). Meanwhile sparse file handling and in-place move and delete operation of archive members have been implemented.
[1] https://lifthrasiir.github.io/aqueous/
Optimizing token usage for my AI coding agent: https://github.com/fluxmoot/fluximp
Its slow going. When I started this near the beginning of the year the ratio between input and cached tokens costs for the interesting open weight models was less than 10 and provider side cache hit rates weren't very good, so it wasn't expensive to break the prefix to remove old things from the context and do rolling summarization to keep the prompt size down.
Now some DS v4.1 Flash providers have a ratio of 33 to 50 and >99% cache hit rates. That makes the cost of prefix cache breaks a lot higher and its definitely harder to get ahead on total cost* of the usual approaches to context management for DS v4.1 Flash specifically.
* - the number of requests is reduced by a factor of ~3 relative to oh-my-pi on the same benchmark; the task is long enough to require compaction once when using oh-my-pi
pg-restic – Postgres backups to S3/B2 (or any S3-compatible storage) using restic, in one small Docker container.
https://github.com/isriram/pg-restic
I needed Postgres backups at work and at home at the same time. I tried a few existing images, but after spending time setting them up I found they shipped an older pg_dump that couldn't back up newer Postgres versions. So I built my own.
It runs pg_dumpall on a cron schedule and streams it straight into a restic repo, so nothing touches local disk. You get encrypted, deduplicated snapshots (hourly dumps of a mostly unchanged DB take very little extra space), retention policies, and daily pruning. If the dump fails, no snapshot is saved. It ships pg_dumpall 18, so it can back up anything from 9.2 to 18.
I now run it in production at work, so I'll be keeping it up to date. I'm no expert, and this is the first time I'm doing something like this (putting something out there), so any feedback is welcome, especially from the database pros.
It's not fancy, but it's something I needed, and boy, I really needed something like this.
I love restic and this project is a truly great application of it! Nicely done!
This is great! Thanks for releasing it. Can I back up my PG to restic's rest server?
Thank you! I honestly wasn't even aware of rest-server. However, I just tested it against rest-server and it works.
Set RESTIC_REPOSITORY=rest:http://your-server:8000/your-repo, put your credentials in RESTIC_REST_USERNAME/RESTIC_REST_PASSWORD (or in the URL), and leave out the AWS variables.
So rest-server seems to have an --append-only mode. If that's enabled, backups still succeed, but the retention step is refused, so each run reports an error and old snapshots are never removed.
I'm building Starkeep, an open-source serverless self-hosting system that lets you keep your own data and build your own apps that run on your devices and your own AWS account.
https://github.com/starkeep-org/starkeep-core
Starkeep's common core provides file indexing, syncing, access control and more, making apps both easy to create and interoperable. Starkeep is easy to deploy, and its fully serverless architecture minimizes running costs. I'm working on an official Starkeep Photos app as well (https://github.com/starkeep-org/starkeep-apps), built for web and Android so far. Everything is MIT licensed.
The first prototype PCBs for my automotive grade, TockOS (Rust) running controller are on order. I've been working on it for a long time, and I'm curious what sort of traps I've let AI run me into. At least for the layout I had to re-do it three times and continuously push the AI to make certain considerations.
Once the design is finalized, I'm wondering what price point I'll actually be able to hit. I originally wanted to be in the $200-$300 price point to be similar to [rusEFI](https://rusefi.com) but I think it'll likely land more around $400. Note that my controller isn't for engine control, just a generic control module, but it DOES actually use AEC qualified parts.
After I'm done with this design I was wondering if I could make a RISC V off the shelf engine control unit, but I'm trying to keep my focus on the TockOS board (Chameleon) first.
Details at https://heisenboard.com
Varios projects all here: https://univerlab.org
working on AudioWeave, an ad free, offline first Android podcast player.
switching from iOS, I wasn't happy with the existing podcast apps, so I built this mostly for myself. I've been using it for the last six months and finally pushed a public beta.
My day to day is BE so I used the project to learn Kotlin and Jetpack Compose. Everything is strictly native using Room and Media3/ExoPlayer. I wanted a nice looking (subjective), privacy-focused app without any tracking or ads.
https://play.google.com/store/apps/details?id=com.enyodigita...
Not available in Canada!!!!
Can you share the apk?
Not available in India :(
Few apps I have originally mainly done for myself; Vaava (https://vaava.app) , been tweaking the e2ee data sync between connected apps, and the webrtc based baby monitor feature stability. Crazy how easily Astra / opus 55 level models create quite complex ad-hoc reproducible (almost automatable) test scenarios with 2 phones, laptop and other tools / peripherals they know they can use, for example for the baby monitor feature Astra did a setup for me which i can reproduce to test that the video + audio works fine between 2 phones for x time.
Also trying to replace own Duolingo usage with Fika Lingo (https://fikalingo.com) basic idea was to first repro the practice of matching words. There is now few other types of language / vocabulary practices. The practice generation works on-device with local LLM, you can give the topic yourself like "Words that relate to Hacker News". Downside at the moment is that English needs to be the counterpart of the practices. Cant guarantee good enough quality with something like germany <-> italy practices. I'm trying to do finetuning with LFM models to see if it would be feasible to improve the local practice generation.
Should I write about the work? I create agent routing. The key is how you define an agent. I define it as a package of memory, skills, and its own scripting tools. Within the deterministic code, it's not the agent that operates but the automation. Assigning responsibility and receiving probabilistic responses is the same as using LLM. However, as you go through life, agents gradually accumulate, eventually reaching dozens or hundreds. It's a waste to include all of this in the system prompt. With our self-developed routing flow, you can automatically select and use only the ones that perfectly match your task.
https://github.com/agentlas-ai/Agentlas-OS
Feedback on your live site, typed or spoken. Each point lands on the element it's about, ready for your team or your coding agent.
https://mostra.design
I'm working on https://hacking.cafe a Unix/Linux environment in the browser.
I was thinking about making Bash an interpreter a library, but I just found that Vercel published Just Bash, which has the same functionality.
I'm working on Eigakan Guide (https://eigakanguide.jp/en), a cinema showtime aggregator for Japan. Every popular cinema site in Japan only allows for very few filters, making it hard to find exactly what I was looking for. So instead I built a site where it's possible to filter by area, cinema, date, time, features like IMAX, English audio, English subtitles, and more. This allows for searching for stuff like IMAX showtimes with English audio on Friday after 7PM.
Since the last time I posted on HN I've added 100s of cinemas, user features like being able to save search filters, movies, and specific showtimes. Also a ChatGPT plugin which allows for searching for movie showtimes using natural language directly inside of ChatGPT (https://chatgpt.com/plugins/plugin_asdk_app_6abfdc298c208191...).
Finally I've added AI search to the site as well, allowing for natural language search with queries like "English audio showtimes near Shinjuku tomorrow" and so on.
I would really appreciate feedback for things to add, things that are hard to understand, and anything else!
I'm not a code guy. I work in medicine. I've been working on a tool that can automate part of my work. If it works well it can free doctor working time and save lives too. It's related to ECG interpretation.
A personal AI librarian that makes large collections of local epub's and pdf's accessible again. It helps people rediscover forgotten files, find information by meaning, explore unexpected connections, and save discoveries for future interests or projects.
Just finished writing my blog post about building autonomous soccer robots: https://hamzah.syedahmed.net/posts/robocup/
Now I'm going to be working on my Linux-subset ABI x86-64 kernel [0], specifically adding support for tracing, then working on allocators. It's written in Rust so I might have to figure out how to make allocation fallible like the Linux kernel does. Also this raises questions about approaches to overcommitting and OOM handling, which I feel Linux does badly, but when constrained to Linux syscall ABI there doesn't seem like many good options.
Also I am not using LLMs for this since this is a learning project, with near-zero real world utility -- I just want to understand OS kernels as best as possible.
[0]: https://github.com/debater-coder/benchix
My main side project at the moment is an AI workspace that's fully customizable.
I've done 80-90% of my work from inside it the past two weeks and made it open source, but haven't made much noise about it yet.
https://lynx.railcode.dev
I'm working on Fraktur, a DJ mixing application that (also) works in the browser: https://greenvitriol.com/posts/fraktur
All Voice-to-text pipelines for linux sucked, so I made my own. I also patched upstream libraries like ONNX to actually work with latest models (was a PITA to make parakeet work!)
It's now the fastest and most accurate linux transcriber.
(fully FOSS of course)
http://voiceio.dev
The learning part seems nice. But I am confused from the small description I read on the website.
What's my involvement in its learning process? Do I have to provide fixes for words so that it knows something was mistranslated first?
No, you don't! It's fully automatic training based on your local voice clips. Nightly or Weekly, when your PC is idle, it uses a big fat model to transcribe your clips, learn, and train the faster smaller everyday model.
You CAN do manual training and corrections; it's helpful, and the hooks are there, but totally unnecessary. I don't do it, it's good enough as is
Working on ZShot [1]. I'm about a year in on a local-only browser tuned specifically for CLI and API usage - kinda filling the gap for "geez, I sure wish I had a browser, but not one to look at". You need a webpage as PNG, or an MP4, or a website as a WACZ or SQLite database. There's a Mac app to help with discoverability of options, but I wanted for my local tooling and building services on top of it.
Along the way I ended up adding an internal agent framework to satisfy stuff like "log in, click the button, and then take a screenshot", out of which fell stuff like automatically generating Playwright tests from that. It even runs the Playwright test suite and Puppeteer test suites ( a bit faster, and for a lot less memory ). I kinda love this bit because I can layer in multiple models that will bounce control of the web session between each other, to do stuff like let a Typesafe model choose to escalate to Opus when it needs higher minded behavior. It's fun to watch it happen.
I've mentioned it a few times around here, I'll probably do a Show HN soon when I feel like it's ready for that. Hopefully soon.
[1] https://zshot-app.com/
Two main side projects lately:
Thesoapbox.email - a social experiment email lottery to see what people do when suddenly given enormous reach. Every day a new subscriber is drawn to email the whole list. It’s hit 10,000 views in a couple of weeks which is my most viral project so far.
The second - a personal AI learning platform built on recursive self improvement and tailored content to push my boundaries. I’m a lifelong learner and this unifies everything for me. (Currently not public)
Continuing with work on my weekend hobby project https://colorhunt.lol You get assigned a random colour and have 24 hours to go out and take photos matching it. The game then generates a photo grid from everything you captured.
Modes:
- Solo → 9 photos by yourself(added 2 new modes, a palette where you are assigned 3 colours and a rainbow where you get 9 colours) - 1v1 → compete against a friend, combined grid at the end - Squad → everyone contributes to a 20-photo grid(added a room option where you can take a break, react to photos uploaded or just join as a spectator)
No accounts, no app install, no personal data stored. Photos and generated grids auto-delete after 24 hours from Cloudflare R2 storage.
Too bad someone stole my MacBook just after adding the palette mode. Waiting for M6 to start selling then resume dev work once I set up the new laptop.
Comments and feedback appropriated
An opensource personal expense tracker, I was a happy user from an offline expense tracker which wiped my data when playstore decided auto-updating it,
I'm at the point where I'm finally actively using it to register my activity, it is straightforward to register tickets by sending them to a telegram bot, then, reconcile those by uploading my bank statement.
I'm trying to get the lowest possible usage friction so that I keep using it.
So far the better built pieces related to the data ingestion. Once this is good enough I'll build reports + anything coming to my head as a useful piece.
I'm leveraging AWS textract + bedrock + bda to parse bank statements. I expected this would have been easier but it is still a very complex problem which had required me to put many iterations (so far it works nicely with my own statements)
For ticket parsing, AWS textract + parsing rules has been enough.
I wrote a bit more on my blog https://alexitc.com/blog/2026-10-16-opensourcing-my-expense-...
Repo at https://github.com/AlexITC/expense-tracker/
I've been working on Heaven.dev for the past couple months.
https://heaven.dev/
It's a cloud where you get completely persistent Linux computers that share a group of resources (CPU, RAM, Storage). This means that you're able to allocate resources across workloads which has a ton of benefits when working with agents.
I found that current cloud infra was extremely restrictive with the new tasks & ways of working that coding agents open. AI is going to continue to change the way software is being built, and I believe that the infra we work with should also be malleable enough to give AI freedom around computers and resources.
The main principle I've had behind it is transmutability -- so computers and resources can be shaped to match any workload.
As every programmer in crisis, I'm making a budget android app. It automatically saves payments without logging into any banks though. https://play.google.com/store/apps/details?id=pl.geckobudget
A friend of mine recently expressed interest in writing a story with me, which gave me the drive I needed to pick back up a project I abandoned a few years back. It's a website for writing a story one block at a time, branching off wherever you feel like you'd want to move the story. I've also used it as an excuse to play with Claude a bit and fiddle with ideas.
Away from the keyboard, I've picked up acrylic painting. I signed up for classes and have been enamored with it, so I purchased the most basic of supplies to paint at home. It's been an absolute blast, and there's a sort of thrill to being bad at something and still finding joy in it.
I know how hard it is to distribute your product and find developers at scale. So I've been busy creating a solution for early stage companies who are low and budget but still want get in front of developers and increase awareness to their product. Hopefully it can help others. daily.dev for Startups: https://daily.dev/startups/
Allegedly, token prices may go up soon, so I'm trying to use tokens more efficiently on my $100 sub.
Noticed that browser agents are not efficient with tokens and also kinda slow. I’m familiat with the web, so I made https://bopbop.dev to control browser and do same things faster and with less tokens.
It is a Chrome extension that lets Claude Code, Codex or other agents use your browser. Comes with a skill/mcp to connect to it.
I used it every day for a month. Now I use it instead of Claude in Chrome.
I made it public last week so it seems fresh but I’ve been cooking for some time
I have been playing with benchmarking agents in a research loop. The experiment leans on using these models to understand and explore builds/gameplay through reasoning game AI over traditional reinforcement learning.
Also, creating interactive ways to view the experiment and visualizating the results.
https://crux.lakin.dev/eval
https://opentaint.org/blog/sast-is-dead/
A data-flow-based, lightweight formal verification engine for application security. Basically, it's SAST++:
1. Deep enough that you really have some formal guarantees about the codebase (like zero findings actually means something, not just that the tool is broken).
2. Fast enough that you can shift the formal verification phase left and apply it to every feature during development (for example, during agentic development sessions to get fixes done within the same feature context).
3. Expressive and customisable enough that you can write specs for most vulnerability types from the OWASP Top 10 (even Broken Access Control and business logic flaws). Skills for writing specs included.
Here is the repo: https://github.com/seqra/opentaint/
https://asm.build - the package manager for assembly projects.
the plot: Programming languages existed as an abstraction for humans. As writing code becomes cheaper, we can write it for each architecture as long as the specification is done well. The costs of running will matter more in relation. So don’t make your AI write rust, but assembly
I've been building https://botlark.com/, which is sort of a HN for autonomous agents. The skill is on the site and on ClawHub if you'd like to take a look. Feedback very encouraged.
AI-first cookieless EU-hosted web analytics. Give your agents access to your site's traffic, client & server errors, and page speed. One script tag install, no cookie banner, free trial.
Started off as a side project monitoring traffic across a network of websites & API/MCP endpoints. We run our own custom CDN ensuring no data at rest in the US or at a US company.
https://fiveb.ar
Too many things.
- See my GitHub [0]
- A terminal workflow tool [1]
You can always see my current projects on my GitHub profile README, I’m pretty proud of it and get great feedback on it regularly. I like to think I don’t build standard crap.
I’m also working on an interactive terminal workflow tool, it’s not quite ready for the world.
It’s effectively an interactive command cheat sheet with its own built in shell, and compostable workflow recipes, all backed by markdown. So your commands and recipes can be read by humans, and executed by anyone (or anything).
Warnings or Cautions in markdown become warnings and cautions when running commands in the shell.
[0] - https://github.com/con-dog
[1] - https://github.com/con-dog/clingon
1-year part-time experiment: can you build a large project with AI writing the code and AI plus deterministic tools verifying/guiding/enforcing it for structural and behavior consistency, without a human reading the code?
I'm building a polyglot web framework in 15 languages, with C# and Lean4 as the reference implementations. Structural and behavioral audits using deterministic tools check that every port stays consistent with the reference. It's about 1.8M lines now. I have not reviewed the code/neither have i verified the behavior tests except at the time of giving requirement and never seen some of the language code at all.
The question is how far can we produce "working code → correct code [Possible] → good code [Unknown]" with deterministic audits correcting structure, ai mistakes and behavior validation.
So far the quality looks acceptable with lots of correction of AI output done without looking, but it isn't fully verified.
Line counts per language, with changes since August: https://claude.ai/artifact/5f1m1pTM7tZ4boQzwsoAG7
the link got deleted accidently and i had my noprocast flag on and hence go locked out. The new link is https://claude.ai/artifact/GaHoaHxqzBiYxoMQYabmJt
I'm bootstrapping a Dyson sphere https://muni.works/research/vision
I use hypertext to map reasoning and arguments. I have AI skills to map out a tree of arguments, all of which are given an initial "debatable" rating on truthfulness. I have a UI in which a human can dig through the tree and mark statements on a 1-5 scale of false/dubious/debatable/plausible/true. Last night I implemented 3 things I identified 5 days previously as potential improvements. https://writing.brlewis.com/posts/2026/my-ma-ballot/
30 years ago I learned to program by building games in Cocoa / Stagecast Creator. I recently partnered with Dave Smith (one of the original authors + members of Apple's Advanced Research group in the 90s, and low-key the inventor of the icon) and we're rebuilding it for the web:
https://www.codako.org/
I'm user testing it this fall with game-building jams in local libraries. Obviously you could build games with Claude, but I think programming-by-demonstration is a great way to introduce fundamentals, and the simulation environment has a low floor and a high ceiling for creativity.
A self hosted software forge / factory / agent platform / mini heroku thing. It's not polished yet but it's building and maintaining little apps and games. https://github.com/flokkit/apptown
You can try it out on mac or Linux in a temp container.
It's self hostable and can use local models. It's not intended to ticket crush at massive scale or produce huge software overnight. It's for deploying smallish things in a predictable & consistent way, with low friction. Then help maintain them long term (e.g. take care of dependency and container updates and similar chores).
SuperBlocker: https://superblocker.app (https://apps.apple.com/us/app/superblocker-ad-blocker/id6799...)
SuperBlocker is a native blocker that works on iPhone and iPad using Apple's latest URL filtering technology that blocks URLs without revealing it to Apple's relays or SuperBlocker's servers.
JSON Query: https://jsonquery.app (https://apps.apple.com/us/app/json-query/id953006734)
The original JSON Query app on the Mac App Store. Store, organize and query all of your JSON with jq or natural language. All of your data stays local so sensitive JSON can stay on your machine.
How does this compare to uBlock Origin Lite?
uOL is a Safari content blocker that only works for Safari on iOS/iPadOS/macOS
SuperBlocker uses a different system that can work across apps including Safari and other browsers.
Demo of SuperBlocker blocking ads in the weather channel app: https://superblocker.app/demo
Outside of my day job, I've been working on:
- hermitator, a tool to run multiple isolated AI agents for WhatsApp on a single machine, with idle agents using no RAM: https://www.encona.com/posts/hermitator/
- showntell, which allows your coding agent to explain things to you with a video that starts playing within a few seconds of you asking: https://www.encona.com/posts/showntell/
- a philosophy tutor (testing in private for now) that does 2:1 live philosophy lessons, each exploring a single topic (aimed at ~10yo kids). I've tested this successfully (i.e. two kids did an hour and they want to do it again), and am trying to make improvements.
- some other education-related things
- a jmail-style mailbox viewer: https://www.encona.com/posts/mailbox-viewer/
Oh, showntell is nice! I'll try incorporating it into my project (obeya.si), which already uses demo videos to show me work it did. But when I ask a question, the response is still just text. Could be so much better with your solution.
A fully FOSS Church management platform called Churchit ( https://Churchit.app ). Jesus once said "Freely you have received; even so freely give". Churchit aims to be a comprehensive and feature rich all-in-one solution that churches can self-host with minimal technical expertise. Churchit provides churches with a dynamic public website, a member portal, and a feature-rich management backend.
I dusted off my 15-year-old game and polished it a bit with AI - https://zabiyaka.com/games/bubblemix/
Despite looking simple, it's fairly well balanced and a quick endorphin fix :)
This is cool, getting early agar.io vibes but way cuter.
Pagecord! The best personal website and blogging platform made in Edinburgh you’ll ever use!
All the features you could want. Open source.
Please kick the tyres, I have an acquisition funnel problem! :)
https://pagecord.com
https://allyourwords.com
a mac app that listens to your microphone and gives feedback on your pace, filler words etc in real time. Built for myself as I tend to get carried away in meetings. Looking forward to feedback if anyone else has the need for something like this.
Just launched https://otusanalytics.com. It's a European Google Analytics alternative with strong privacy options and modern features such as an AI assistant that can act as your web analyst
I've been building an art therapy sandbox upon Pavel Dobryakov's fluid sim, and I'm almost finished, and going to release on Steam, and shortly after, I'll be releasing it on GitHub fully open source.
Over the years I've used it myself and have found it very soothing, slowly adding to what it can do. Hopefully others can find it soothing and beneficial, just as I have.
You can play with it in the browser for free at http://swirltogether.com/
(semi-related, I'm seeking design / design engineer work, I'd love to be pointed in the right direction if anyone knows of a role that might fit, my email is in my profile)
I've been working out some of the math behind knitting stitches and made a little write up linked here:
https://asunder.co/app/knit/articles/klein-four-group
It's quite basic but figuring out the real-world bridge between algebra and knitting and re-discovering obscure stitches is quite satisfying.
Putting my applied math degree to use for once :)
For the past few weeks, I have spent my hobby as building and improving my own coding harness, VT Code. I plan to keep improving it throughout this month. Recently, I built a native macOS using GPUI (gpui.rs). Docs: https://github.com/vinhnx/VTCode/blob/main/docs/development/.... I'm pretty happy with how it's turning out, though there's still plenty to improve
I’m finally doing the thing I always said one day I would, and writing a game! Early days yet, but I have proof of concept for much of the rendering tech now.
https://fluff.toothycat.net/@moonshadow
I was working on platform to build 3D room design for everyone who loves interior design, no need to install, no need to pay. You can see my published design here : https://formanest.app/p/2LxQNrjg50vvxrbb
I'm working on a Markdown editor for writers & researchers (https://grasmere.app). A bit like a cross between Obsidian and Scrivener. I'm finding the marketing part hard but slowly getting there
Continuing to plough my own furrow with Opair, the coding harness that eschews autonomy:
https://www.opairdev.org/
Also hacking away at a little encrypted messaging / key management app, where all the crypto happens client-side so your private keys and unencrypted message text never leave your device:
https://www.pgpal.org/
I've been working on a project to reduce attack surface of Linux document readers, starting with Okular: https://github.com/l4rzy/okular-mupdf-ng
The idea is to sandbox the most dangerous parts of it: The Document parser and renderer.
I have been in enterprise for couple decades now, learnt and experienced the challenges/gaps between the specs vs implementation, the asks vs the real needs in the space of identity, security and access
Product: webfrolic.com
Demo: demo.webfrolic.com
What it is?
Consider Okta (Workforce identity)+ Auth0 (Customer Identity), Authzed (authorization) + ServiceNow (IT service Desk) + Zendesk (Customer Support Desk) architected together
Being an engineer there is this inherent habit of prioritizing correctness over marketing so I would not oversell it, though its probably the most secure and complete thing that could be built from identity perspective
Status (WIP):
1. I am still debating/settling on how to incorporate the AI Agents needs into the architecture, big focus on avoiding the creation of security or performance foot guns where administrators leave room open for abuse with configuration despite all this attempt to build secure system
2. Feature stress testing and performance testing
Hence (https://hencegame.com) is a modern twist on Einstein's puzzle, with fun picture clues instead of text. It's a free online daily puzzle, the same for everyone, where every puzzle is always solvable by pure deduction with no guessing. I launched it October 5. The hardest part was the graphics, making each picture clue unique enough that you recognize it immediately and know what category it's from, even at small sizes. Easier said than done.
Every puzzle is run through a verifier first to make sure it is solvable without any guessing, just by pure logic, so there is always a next step to find, which is what makes it feel so rewarding. There's a hint system to help and it works really well if I do say so myself . No account needed to play.
From what I've been hearing so far, the people who have checked it out are enjoying it and can't wait for tomorrow's puzzle, so that's a good sign. I'd love to hear how the difficulty feels, and whether the pictures read clearly.
I've been working on a multiplayer card/board game with some friends! We've been using Unity (C#) and its services like Relay and Lobby. Working with Unity has been pretty solid so far but I cannot recommend Relay and Lobby (now multiplayer tools) as their docs are missing a lot of crucial details such as how to preform host migrations.
We are hoping people who enjoy Magic/Hearthstone will appreciate the complexity and mind-bending nature of this game:
https://store.steampowered.com/app/2865190/The_Time_Game/
For now Im working on an agent-orchestration application that is taking tickets from a tracker and giving them to an Agent. It's called Even Keel.
The main feature is, that I can use different subscriptions/API keys (Claude and Codex - v0.1.4) and trackers (GitHub, Linear and Huly - v0.1.4) to make Even Keel start an agent for every single issue, task by task, one at a time per folder. There is also a parallel feature, so I can run more than 1 project at a time.
Once upon a time, when codex agent was integrated into the app, I wrote a ticket to make codex his own hook. It reused a hook that blocked its own git merge, if tests did not pass. I started calling it "A new agent writes its own cage"... Right after, it turned out it doesn't listen to the hook and it fixed it
Version 0.1.4 is already shipped, but v0.2 is coming soon with more supported trackers, agents and new features.
https://github.com/NamerekT/even-keel
I've been working with orchardist and gardening groups to build an app to plot and track your trees over time. It's useful to be able to look back and see when you graft trees, or when they fruited over time.
https://treesmith.app/
I’ve always wanted an interactive history timeline, so I tried to create one [0], but I am not happy with the result. It’s difficult! [0]: https://jonasfoyn.com/timeline
A slow, realtime strategy game similar to Neptunes Pride and Subterfuge.
Each game tick is an hour and games can last over a week. I got hooked on both Subterfuge and Neptunes Pride and always wished there were some more titles in this genre.
Currently in closed alpha.
https://ashdomes.com/
Working on browser games https://games.numerical.works/
I've been playing bossa nova guitar for many years and I've always wanted to share the chords of some of the songs I've learned by ear in a fun way. Since I have zero front-end skills and little time I was never able to. Now with the help of a model I created a nice webapp that syncs the chords with their youtube videos (the chords were all manually added, only the UI was vibecoded). It's been fun and a great way of learning the names of more complex voicings.
https://chords.pacuna.dev/songs/desafinado
I love this! Aguas de Marco is one of the most beautiful songs I've ever learned. Definitely going to learn some more bossa nova chords with this!
Rebuilding my ERP to work more like the model of https://plaintextaccounting.org, where I expanded the idea of `double entry accounting` to `double entry trade between parties`, plus several other ideas for what I see a way to collapse a lot of complications that are typically for ERP systems.
I make a special `textual DSL` where I model business scenarios like `Make a sale that include 3 invoices : 2 to different cities in country A in currency USD and other in country B in currency EUR` + all the intermediate steps like transportation, manufacturing, etc.
I`m now in the woods with a simplification for the price engine, I dislike much how complicated is normally, and I think I have good enough solution for most cases that generate without dynamic calculations.
Given the popularity of decision models I tried to create a standard text classification pipeline that runs entirely in the browser using webgpu/wasm. Downloads an embedding model, trains the classifier (a Linear SVM for now, whose hyperparams are determined by cross validation over a grid), calibrates the classifier so that reliable confidence scores may be produced. This is in Python that is run using Pyodide. You can paste your own data (no uploads supported) to train a model on it. WIP, and mostly generated using GPT Sol. Testing a few things out. Will have a blog post soon.
https://pocket-text-lab.abhishek-ghose-82.workers.dev/
I’m working on Spool, an open-source tool for running persistent Claude Code loops.
https://github.com/enes-alatas/spool
I wanted a way to manage multiple loops without babysitting terminal sessions and to talk to them as they work. With Spool, you can create and manage loops from a web UI, chat with them through Telegram or Slack, and let them communicate with each other. Each loop can run on your environment directly having access to everything you have, or in its own container or even better on Anthropic's cloud.
I’m using Spool to build Spool and figuring out what works well when several loops collaborate on the same project. That's also an interesting experiment going on.
I’d also love to hear how others are managing long-running or persistent agents.
I am working on https://wateaminbox.com and https://learn-thai.ygncodelab.com. I got trial users who tried wateaminbox and some random people message and thanks for wateaminbox.
Planning to add more channels to wateaminbox and learning how to get paid customers.
Generate subtitles and dubs for any video, in the actors' own voices
https://github.com/swax/dub-anything
I use it to watch movies that have never been translated to English. A whole world of previously inaccessible media is opened up.
Oh, with the arrival of LLMs, I actually have a few projects!
My homepage has gotten a dustoff and is up at:
https://hypertekst.net
I'm volunteering and running meetings - and I wondered how tricky it would actually be to get decent meeting minutes and record a (small) multi-person in-person meeting.
Turns out there's a few tricks to recording a room on Android - and rather tricky to get good meeting notes (in Norwegian) from the audio using only local LLMs - and avoiding the data touching the cloud at any point.
I need to revisit Android recording app recommendations - and what combination of incantations will allow you to transfer an audio file directly to a MacOS laptop (apparently it's not quickshare and airdrop unless your Mac is wide open...).
But I have a wip single download app that lets one record and transcribe - should work on MacOS (tested), Windows (not yet tested), Linux (tested on chromeOS/Terminal).
The UI is a vibecoded mess:
https://hypertekst.net/steno/
I'm also working on a forum/chat app - cross section between zulip, wiki - Google wave and discord (text chat). Still pre-alpha.
And a character sheet/generator for ttrpg - oath hammer (closed alpha testing) - a follow up to my oath hammer dice WPA:
https://hypertekst.net/dice/
And it's bloated 3d cousin:
https://hypertekst.net/dice_3d/
And a campaign managing system and sort of a illustration experiment with 3d scenes and LLMs...
Ed: and I just realized you can use a Google form to create a Google form... So I set up a small script to generate meeting/event sign up forms...
What might be of interest here is the launch video I made for my app, Fit Shapes—combination of Claude turning the app screens into a video, GPT Image 2.5 for image stills, Fal.ai for animating the stills and Elleven Labs for music. An interesting thing about agents creating music is that they can create and adjust tracks in segments to bring in specific elements as needed.
I know there are a thousand of macro trackers out there, but most of them are some combination of ugly, tedious, expensive. Since it's an app I use every day I tried to make my own with a focus on removing taps.
Anyways, video is on the homepage.
https://fitshapes.com/
How to make data storage on multiple storage devices unbreakable and consistent? The former means for me that back-ups should be automatical and serverless, the latter means that no matter how much of that HDD/SDD devices are connected, I want always to know all metadata of all files I have. I want this software under GPLv3 if to consider GPL still protects me from anything because permissive licenses are no better than WTFPL by any single case.
My first idea is to require the user to have more than one disk for the system needs and the system should know all of the rest data devices user owns. My second idea is to treat different files as having different priority which impose multiple copies. My third idea is to make the software accessible, and the last idea is to make encrypting of all the devices easy and with no ties to any commertial entities. I would like to have a discussion about how to not lose anything digital.
It's a bit stupid but I made a free horoscope api during covid and hosted on vercel, it started getting a lot more traction since and I have migrated to a proper domain and to cloudflare workers. Getting around half a million request / day and bill is just below $10
https://www.freehoroscopeapi.com/
A couple of things, but mainly:
- Scripture: https://scripture.wolftrail.net/ - a website I use to daily read a couple of Bible verses according to the Russian Orthodox tradition. It's still not fully functional, but useful for that purpose at least.
- MacGames: https://macgames.wolftrail.net/ - reverse engineered vintage Macintosh games (System 7 era) that can be played in the browser.
Been working on my version of a journal app that is basically Diarium but I can add any features that are missing from Diarium and Day One. Been a cool side project.
I wanted my journal to have a proper “writing” mode where everything disappears and it’s just the words.
Better voice to text transcriptions. Ability to paste the transcription into the body of the entry but also keep the recording.
Better picture layout.
Apple Watch integration so I can just speak into the watch when I have an idea or something I just want to say.
And when I think of something new it’s very easy to integrate. Love it.
For the past two years I've been working on my day planner app Ember.
Most planners treat your schedule as fixed, so the moment a meeting runs long or a task takes twice as long, the plan is wrong and you either fix it by hand or give up on it. In Ember, if you move a task or change how long it takes, everything after it shifts later, and it flows around fixed things like meetings. It's been super helpful and I have been adding new features ever since.
Right now I'm working on a native Kotlin app to replace the current one is web + Capacitor and exploring full duplex voice models so I can add or move things around the app while having a conversation with an agent.
homepage: https://emberflow.space demo: https://demo.emberflow.space/?ref=hn
Software for families:
https://santawho.com A funny holiday game you play in person with the whole family on just one phone (the parent interviews everyone for jobs on the North Pole and surfaces inner insights about what everyone might enjoy from Santa). It’s for any age, private, ads free, just pure holiday spirit.
https://www.dreamlist.com - collaborative shopping lists and online wishlists and gift drive platform. Again ads-free, private, stack multiple lists into groups of lists to share on social media or with grandparents.
A number of other ones about to launch as well. All little tech, not big tech - private by default, no ads, no feeds or media, just surfacing what is needed.
GeoPeek, a small macOS Quick Look extension for GeoJSON. I work with geospatial data and wanted to press Space on a .geojson in Finder and see a map instead of raw JSON. It also shows an attribute table and basic stats. https://github.com/serifsadi/geopeek
ooh I like this. I deal with a lot of random geojson files in my own project, https://landrecords.us
VibeVoted (https://vibevoted.com/) - a platform for product assessment and competitions based on the rankings of 10 leading AI models from 10 different labs. Also, each product get's feedback and recommendation for improvements. Here's a sample voting session: https://vibevoted.com/s/password-managers-2026-oct
A GEGL/babl based image raster/vector/pixel editor for Mac. Using metal for all effect on-canvas previews to both have speed and full xcf compatibility.
An ide that has a full agentic plugin that allows ablating functions or pieces of code that one has to complete. So instead of leet code you can build towards a program and still learn something. It can also guide you with bug fixes and other stuff. And it also includes a fully integrated dos environment with a fully functioning debugger and asset editors. Also works on iPad.
I am working on a Surf app for Nordic countries, the wave layers presented by the large forecast groups are too course to estimate good waves based of local winds in very complex terrain. No public demo yet. If you have any experience working with weather apps pls link some ty
I'm getting hit constantly by bots trying to hack my site, so I'm trying to hack my own sites. Pretty much I have a few obliterated LLMs (DeepSeek-V4.1-Flash-Obliterated, OrcaCyber-Zero, etc) which run on their own sandbox to try to hack me.
If anyone wants to try it, you can test it on mockhack.com
https://www.abejora.com - time tracking for freelancers and small teams
Lately we've been working on access for AI agents. A lot of people want their own agent to do the repetitive part of time tracking. Control remains important, which is why we rebuilt our core this month to support change history and full undo/redo functionality. Feel free to try it with your own agent.
We built this after the timesheet tool we paid for raised its prices and moved some features that we used regularly behind more expensive plans. We had to add three new colleagues, and everything combined this would cost us $500 a year more for the same software.
We run Elixir and Phoenix LiveView on a Kubernetes cluster made of Talos instances.
We're still looking for feedback on the product, so please let us know what you think!
We also added a live-demo functionality to make it easier to try. It spins up a workspace just for you, with sample data in it. You can make that workspace your own by adding your own data and undoing the creation of the sample data via "History". It should just take a single click.
Anyone working on something for Flipper One or Mecha Comet? Anything interesting related to x402 or cryptocurrency in general?
https://trymeli.app
Meli: A spending tracker for Mac designed with privacy in mind. Turns credit card alert emails into spending insights and budgets. I was in security for 15+ years and I built it because I didn't want to use Plaid or other aggregators and wanted my financial data locally on my device and not in the cloud. Spreadsheets required a lot of work to maintain so I wanted some sort of automation. Connects directly to your email (reads only alerts from supported senders) and all parsing happens on your device.
Works with seven major US banks (AMEX, Capital One, Chase, Citi, Discover, Bank of America and Wells Fargo) plus PayPal and Venmo.
Does this list cover most of your spending? If not, which banks or cards would you need?
Like 2h ago I was doing some research on that topic. I’ve heard about open banking a while back and thought: Oh that’s easy. I’ll just get an api key from my bank or “oath” into it. Sadly it’s not that easy.
How do you get programmatic access to your data?
Unfortunately it's not easy. In the US specifically banks don't generally offer APIs or keys to access them to individuals and you have to go through an aggregator like Plaid (which is what I want to avoid).
However, most banks do send alert emails when something is charged to a credit/debit card. Meli connects to your inbox read-only via OAuth (Google and Microsoft) or IMAP for other providers and reads those emails on your Mac.
Of course there are limitations to this approach, you only get access to the data that is available in those emails like merchant, amount, date and account.
Which banks are you interested in? Are you based in the US?
At work, I'm setting up a harness to make research, experiments and doc work more efficient, and at home, with the goal of working with a game engine, I'm trying out various tools besides ChatGPT and Claude, like music, sound effect, video generation and motion generation. What I feel is that even the companies working on frontier models still don't seem to do well with sound that isn't voice (songs, sound effects) and motion understanding (not video, but things like moving a character) (though they might have it internally and just not be releasing it).
I couldn't find an open-source inventory management system I liked, so I started working on one. It was designed to handle multiple stores, each with their own inventory and products. Each product in categories and sub-categories, and location management for each item. It was supposed to be the base of a few online stores and a brick and mortar shop, but that business plan fell through. Still working on the inventory system, and it still needs more work.
https://github.com/SeleneSoftware/inventory
Built an app to suggest best local beach for given weather for my town. My partner shared with local moms and they seem to be loving it - https://beach.metacursion.com/
Also revived old project - https://aminosplore.metacursion.com - the idea was to compare what essential amino acids each food contains using spider web chart. It’s been dead for a while due to USDA API changes so I revived, added macros and enabled to build foods into actual recipes.
Finally have been improving elsrc.com - an easy and scalable way to use ai image generators. If you need hundreds of images you can just request them by url. Supports versioning, batching, regional and national styling based on viewers IP and few other neat features!
My co founder and i (aka Opus 5.5 and me) have been building https://www.dexbindr.com/ for the past few months now. It started with wanting to complete a pokedex binder but i hated carrying my 11 pounder card binder to card shows. Decided to build this app to help with card pricing, scanning and tracking to get my binder to completion. And apparently people actually do find this useful enough to pay for it so another example where solving your own problem can also help others.
Working on a GitHub App that reviews PRs. https://pacecar.sh
The bottleneck in most AI coding workflows is the actual PR review, so this just comments on open and update, including drafts. I tried a bunch of setups until it was cheap enough to leave on and still fast enough that you're not sitting around waiting.
A fast macOS-native Bible search app
- lays out all 31k verses in <10ms in a single scrollable window (tested on my M1 Pro MacBook)
- each keystroke in the search bar filters results and lays them out in <6ms
- deep functionality in search syntax (ranges, quoting, etc.)
- click an English word to search for verses containing the Greek, Hebrew, or Aramaic word that the word translates
- given a Greek word, matches OT verses where the Septuagint has that word
- built-in Greek and Hebrew lexicons, displays entries for each original language term in the search
I'm working on an open source agent and engine that can calculate your Federal and California taxes. My goal is to build an AI system that can act as your CPA, and hopefully allow you to do super cheap tax estimates and planning.
Looking for help as well!
https://github.com/captaincole/OpenReturn
I’m building Spine, a platform for agentic applications because I want agentic applications to feel more like traditional software: something you can reason about, inspect, reuse, and distribute.
https://0xc0ff.ee/posts/im-building-spine
I try to 'vibe code' a product every 2 weeks. Most of them go unpublished, I just build it for the sake of building it. Most recent project is Unmute Party (https://unmuteparty.com) - multiplayer party games for remote teams, played in the browser during a Zoom or Teams call. The server is EU-only for now, and I plan to add US servers next week. Most of it is free, the paid tiers aren't even working yet. The paid tiers can be unlocked by using code HNWHATAREYOUWORKINGON.
Apart from that, running a travel agency I'd never be able to without agents (they manage my ads, social media, website content etc.). Started with boa.ba and it went quite well, now preparing dobosnia.com
I'm building Rank - a language I could comfortably write on my phone. I'm even thinking about making a dedicated handheld with a physical QWERTY keyboard and multi-function keys like the ZX Spectrum had
https://rank.viewflow.io/
The language combines ideas from BASIC and APL: array operations, words instead of special symbols, and data-first function calls
FreeBSD 15.2! Scheduled for release in early December. Biggest change in 15.2 from my (release engineering) perspective is that I'm using got instead of git, dramatically reducing the dependency tree.
Git got got
I'm working on my first hardware device, based on ESP32, it's an AI employee in a cube.
https://elestio.geta.team/web/getacube/
Column view for hacker news with resizable fonts. It’s just a fun hobby thing for me. I’m not in software, but I can do hobby level programming and that gets leveled up by codex.
https://hn.orbitalchicken.com
Writing procedurally generated civilisation, mostly inspired by Crusaders kings mechanics, with personalities driving most of historical events. Haven't had so much fun in a while
I wrote a children’s story in Galician about open source. It's told as a fable. I made it open source, and people are now helping translate it into Spanish and English.
I’m looking for ways to fund printing and distribute the book. If anyone has done something similar, I’d love to learn from your experience!
Visor. It’s a new type of trading terminal centered around options and gex. The main feature is overlaying the options payout graph right on the chart.
Think tastytrade rotated 90° and put on your live barchart and then watching all the Greeks play out in real time.
Has been instrumental in me deeply understanding the dynamics of more advanced options trades.
I'm working on a table tennis VR game with human-like AI. It's trained on roughly 25 hours of me playing against myself and runs in real time on a Quest 3: https://loopspin.gg
I am working on https://github.com/fastagent-sh/fastagent
Which let me deploy pi-durable agents to AWS, Cloudflare, Fly.io as services with every cheap cost(in MicroVM)
What I assume that the agents should run as a "Web Service" on the cloud and do the work silently and complete the work without the chat. We don't chat with our system monitoring, don't chat with log system.
https://lars147.github.io/foldroute/
FoldRoute - a PWA route planer for folding bikes & public transport
I bought a Brompton a few months ago and realized there is no good planer for combining biking and public transport. I one-shotted the first approach with codex. It was so decent that I wanted to elaborate on the idea. Still not perfect and needs a lot of testing in the wild (especially the navigation). Currently, only in German. If there is demand, I will translate to English. For now only a PWA. Therefore, it has some limitations. Maybe I will make it an Android / iOS app one day.
I am working on an open source and local app for tracking one’s wealth.
I wanted something without ads, subscriptions and cloud, where I could throw in all my accounts, stock and crypto positions, loans, funds, estates etc to have it stored and tracked on one place and via MCP exposed to my Claude financial advisor.
https://moonyfinance.com/
I’m working on https://greenflagged.dev
I originally thought it would be useful to have a package registry that helps prevent supply chain attacks. It reads and judges all of the changes to popular NPM packages. While building I realized that it’s a package manager that agents can settle package versions with as old versions have new vulnerabilities found.
I’ve caught several supply chain attacks since I started monitoring, and am building an automatic response pipeline to validate and publish new attacks.
I’ve also been working on python support and hope to have that released soon.
Working on modern, self custodial money movement. Especially relevant given recent Ueno Bank collapse in Paraguay.
https://peanut.me
https://www.rgbloom.com/
RGBloom: create printed photo books in minutes (UK only for now) for just £10 including delivery. Been working on this for 8 months, recently launched iOS and Android apps to go alongside the web app. Got just a few dozen orders a week atm.
I'm focused on making it privacy first, keeping as little data as possible for as short a time as possible (no accounts, you don't even need to provide an email address, and if you do it's wiped in 30 days). And transparent pricing: not getting hit with large delivery fees at checkout, and no discounts just honest prices.
Really good idea - hope it does well!
I'm working on 3 products now:
1. FormBeep(https://formbeep.com) at $38 MRR now - A form to Whatsapp tool. Stack is mainly CF workers with Clerk Auth.
2. SlashLink(https://slashlink.app) $0 - A simple pay as you go amazon link localizer. Great for affiliate folks and a cheaper alternative to Genius. Stack is same, CF workers with SQLite(D1)
3. StackStats(https://stackstats.app) $791 - A better advanced analytics app for Substack writers, fully local. I have been writing on Substack for a while now and their analytics suck!
Curious how many of these comments get even a scroll by. There are 500 now.
I hope this isn't the only way you guys are promoting your projects, assuming that you want to drive some traffic.
I'm building an open source quantitative finance pricing platform built on top of QuantLib.
Think of it as a distributed QuantLib, exposing pricing, curve construction, risk analytics... through APIs, making QuantLib easy to integrate with applications without having to deal with the C++ library directly and an extra layer of abstraction on top of it.
https://quantra.io/
A networking-over-DNS client and server, in Go. https://sr.ht/~nullenenenen/ShrewDNS/
A bit like iodine in the basic idea, but the execution and network topologies you build with it are different. It's more of a building block; you get your DNS channels represented as UDP endpoints, and take it from there, ie you can run Wireguard over it, or SSH or a SOCKS5 proxy (using quicmux, so over QUIC), or run Mosh straight over DNS if you're crafty.
It has an Android app too! I tried to submit it to F-Droid but they're so swamped in spammy or low-quality apps that they're not taking any brand new apps :-( A sad state of affairs, that.
I have been working on building an agent and workflow builder that converts chats with claude, chatgpt, gemini or any other chat agent into a workflow or autonomous agent. Its live at toolscaled.com
Also built a free apache 2 agent/ project management system: https://github.com/mhalstrom/maximizepm
Essentially I found the key thing with modern ai assisted development the key thing is to manage your own time/ parallelize without having to switch between chat session as much as possible.
I am working on https://gangway.sh/
I always wanted to quickly deploy small sites and apps for clients and friends.
While developing this, I realized that this also could be ergonomically similar to Claude artifacts for open source harnesses.
I have a few users that I personally know, but hoping this grows to have more users one day. Not getting abandoned any time soon.
Self hostable and easily forkable with a Apache 2 license.
In my free time I'm doing a lot of things, it's a little crazy—an Imperial Rome TTRPG, an archeological research RAG db to fact-check it, a browser-extension agent harness, experimentally training NCM models (eg entity extraction (didn't work), or playing Wesnoth (jury's still out)), US Cartograms (maps with proportionally adjusted areas), and Jug.fyi (a probabilistic distribution calculator).
A NetOps oriented benchmark and a network investigation tool designed for llm to ace that benchmark; which sound wrong I know :)
If you are into agentic SRE or networking tools:
https://cloudywithachanceoflatency.net/blog_04-investigative...
The tool is open source, the benchmark itself should also be published soon)
Cloud save sync between a real Super Nintendo and emulators.
I’ve just got back into hobby electronics after a fifteen year diversion into software. I set up my workbench this week and this is my first project, something I first imagined building twenty years ago!
I spent the weekend wiring the spare controller port to a microcontroller to act as WiFi bridge. I have a flash cart, so I patch the reset vector to a small save sync routine that talks to the microcontroller.
It was especially magical having Claude read from a logic analyser and flash firmware tweaks iteratively. I still got to wield the soldering iron at least.
Adding more features to our SaS offering for parsing documents into structure data. https://www.docuapi.dev/
My 10+ year old SaaS for test prep: https://crucialexams.com/
I recently rebuilt my COVID project, a Steam/online version of the drinking game Kings Cup: https://store.steampowered.com/app/1506610/Kings_Cup_The_onl...
A friend and I have also started working on a restarting/brewery manager game that is not released yet.
https://github.com/Vermeulen-Solutions/baseline
Astro site starter template meant to be used together with agents.
Mostly working on my personal project, pi-container[0], for running the pi coding agent in an isolated and more secure container environment.
[0]: https://github.com/pkhamre/pi-container
I've been doing the same, just with a llama-cpp local serve integrated + tor as a proxy. I was wondering whether there's a way to limit the heating issue with macbooks when running local model: do you think that limiting the number of cpu in the container is enough (albeit slowing down the llama-cpp inference speed) or is there a better way?
A small team and I are working on Sendery [1], a transactional email platform for product teams. We're trying to fix some of the headaches that come with managing product emails as companies scale: keeping track of hundreds of templates, maintaining consistent designs, coordinating changes between developers and marketing/ops, and managing everything across multiple languages is way more time consuming than it should be.
It's been a fun project to work on since it's been a pain point for myself on many projects, so I'm mostly just trying to solve my own set of problems and hoping other people find it useful too.
[1] https://sendery.co
https://getlibrium.app/
I’m working on a document library app for macOS with tagging, collections and user attributes. Unlike other apps that pull files into their own databases or reorganize files into their own folders, Librium indexes files in place, so your existing folder structure stays in tact. It works with folders on your local drive in addition to external and network drives and the cloud. Search works with on-device text extraction and OCR.
It’s currently in Test Flight beta, but I’m planning to have it to App Review this week.
Building Wishfox [1] for around 6 years. It's a wishlist app with price tracking, groups, secret Santa and more. Recently did a major rebranding, and the mobile app is almost ready for testers.
Then I'm working on improving my two open-source apps, Portfox [2] and Localfox [3]. Portfox shows ports for running services and dev servers, Localfox provides them with .localhost domains with TLS and Cloudflare tunnels.
[1] https://wishfox.app [2] https://github.com/xergic/portfox [3] https://github.com/xergic/localfox
I've been slowly working on migrating things to a more well documented setup as I build things. For the longest time I just set stuff up and then.... Hoped for the best. Started with my main authoritative DNS for some of my domains today, still working on some Ansible setups but I have the zones down in Terraform now. Is it overkill? Absolutely. Am I learning stuff? Also yup
I'm working on a cross-file deterministic drift checker https://github.com/mennanov/blockwatch
The idea is that the source code is annotated with special rules which configure how parts of the code (also configs and docs) depend on each other. This can be done either directly in the source code comments or in a config file blockwatch.toml
This tool is very useful for large codebases that e.g. share constants across programming languages and configs, have docs that should strictly reflect the current state of the code, etc
It uses tree-sitter to parse the source code, 35+ languages are supported.
The project is actively developed, your feedback is welcomed! Thanks!
I have been working on marketing tools for SMBs. Text message marketing is quite possibly the strongest but most regulated channel for marketing.
Here is the new site design I'm working on: https://getsetreply2.com/
Here is the existing site which will be updated to the above one: https://getsetreply.com/
Euclase: http://euclase.ai - the creative harness for generative AI
Unifying all the various image/video gen (comfyui for local, api for closed) in one cohesive desktop app, with integrated agents to pilot the whole thing (if you want). It's been fun to build and quite fun to use. Technically out in beta and testable, but working on refining onboarding/docs before I try going wider.
Finding a balance between deep enough to satisfy more technical creatives (like me) and less technical ones. Agent integration helps tremendously with the latter.
Still working away on my book (and library book) recommendations service https://leafle.org
It has if nothing else been very useful for me (it has correctly deduced that I would get something out of borrowing Grief is the Thing With Feathers, the Dispossessed, and Cosmicomics, none of which I would have thought of reading). Plus it has been a fun opportunity to play around with LLMs as an interface and a research and reverse engineering tool; as well as Jev for non-user facing quality work like banning crap openrouter providers based on their output before putting it in front of a user.
Most recently I’ve been dropping off flyers for it at a few local libraries in my area.
I tried it out nd it looks nice!. It’s a great way to support local libraries and bookshops. I hope you have a ton of success!
Thanks! Hope it found you something interesting.
Got tired of postman being such a resource hog, so I made a native app that does much same thing for iOS and Mac.
Eventually reached parity at around 4mb and added iCloud synch and mcp so I can control my test suites from Claude on my Mac.
Making it available for others to use this week after seeing a lot of people are also feeling done with postman
https://testflight.apple.com/join/Xp5qjGsg
Oh fantastic, I've had this on my todo list and now you've built it for me?! Thank you, excited to check it out!
I think a lot of people have variations of this as personal software. I’ve just been tracking on requirements and updating mine. I don’t have grpc and JavaScript scripting but it does no code style assertions they claude does a good job of maintaining.
Let me know if anything glaring is lacking. I was wondering if people want the test suite results to a remote Postgres instead of the local SQLite but not sure it’s needed
This month I'm working on several projects:
1. Broadcast (https://sendbroadcast.net) - Self hosted email marketing software. Been working on this since 2024 and positive revenue. Releasing v3 this week! :)
2. Render Screenshot (https://renderscreenshot.com) - Screenshot API (very much a scratch my own itch project)
3. Widgetful (https://widgetful.com) - A menubar app for managing multiple Claude Code agent sessions
APIKulture. An alternative to Postman/Insomnia/Bruno.
To be honest, I started this because Postman and Insomnia became forced-corporate-cloud not usable for me anymore, and Bruno lacked a feature I considered fundamental for my workflow.
The feature is to remember the JSONPath expressions between requests, a feature that Bruno has implemented already.
But I created this thing and I'm using it daily since then. C++ and wxWidgets, it uses 40MB of RAM, and it is super fast, instead of the 400MB used by the above mentioned programs.
https://www.apikulture.com/
I started working on a little news aggregator which has kinda spiralled into more of an open web thing. Keeping it open source, and open to feedback on how to better improve it.
I’ve tinkered with some decision model stuff too, and it can do things like identify possible spoilers in comments, or spiciness levels in comment sections. Works with HN, Bluesky, Lemmy, Mastodon, and can import RSS feeds too.
https://paperwing.io/
My main projects are still Grasshopper https://addons.mozilla.org/en-US/firefox/addon/grasshopper-u...
And Meltdown https://meltdown.merkoba.com/
Grasshopper is a tab manager for Firefox.
Meltdown is a an AI harness made in python and tkinter.
You can check screenshots of it and descriptions in the link.
Also I've been blogging a lot recently on a weight loss blog. I like asking Kimi for reviews on what I write.
3 products:
- https://polygonjs.com/: a design tool to create interactive 3D configurators
- https://polyreprint.com/: a hosting platform for those configurators. See more enterprise-type examples on https://polygonjs.com/configurators
- https://polyreplay.com/: game made with said design tool
For reasons, my team is pretty much limited to Argo Workflows for our CI.
Nothing against Argo Workflows, but I really wanted to get out of the business of building our own CI product on top of it.
So, I’ve been working on this CI “framework” on top of Hera (the Argo Workflows Python SDK, which is really neat) that will hopefully make it easier for our team to build CI pipelines, with CI “primitives.”
It’s called Stagecraft: https://github.com/elliottmoos/stagecraft
I'm working on building browser-based learning games that help my 6 yo' to explore and understand small concepts through tiny games https://chalkolate.com
I build it based on what I thought daughter could improve and ideas that my daughter gave. I am currently working on introducing basic electronics.
Still plugging away on Slopjective-C 3.0 https://github.com/doublemover/Slopjective-C/pull/8278
and many other things
As I'm working on my regular decision college applications, and since Opus 5.5 released, I was able to polish up Quillium (https://quillium.bryanhu.com) so I can dogfood it for my college applications.
It's a nonlinear writing editor, so I can keep track of all the versions of different words, paragraphs, sentences, or drafts. No more Final_FINAL_v3_essay.docx
Working on a shared folder that syncs for agents, where you mount your existing data, apply controls / permissions and can monitor whats going on in real time: https://www.filestash.app/agents/, the code is in: https://github.com/mickael-kerjean/fdrive
I'm launching https://rallyrama.com, a rally time trial game for iOS and Android with retro graphics and cars from the 90s.
I have zero experience in game development. It's unbelievable how far you can go with Claude these days.
Still plugging away on Flopper to build a holistic view of the GPU market including providers, rental prices, vendor information and mostly - specs for GPUs.
Added a way to look at power/flops and new model requirements lately to try and estimate how many GPUs you need to run a certain model: https://flopper.io/tools/llm-gpu-calculator?model=llama-3-3-...
I've been building RCHIVE [0], a desktop app for cataloging, searching and reorganizing files on your computers and drives.
I've got 30 years worth of poorly-organized junk strewn about various computers, NAS, backups, folders, etc, and every year that passed made it harder and harder to rein it in to something sane. And last year I had one drive start to go out on me, so between all that, and consulting contracts drying up like a shrub in the Gobi, it felt like time to build the tool I always wished I had. Ended up rolling the dice and turning it into a real product too, took about 14 months (mostly full-time) to get from start to launch.
Anyways, it scans many different sources, gives you unified deep search across the entire catalog (drives online or offline), allows you to virtually organize and clean up giant archives of digital stuff, resiliently copies or moves files when you're ready to actually restructure on disk, or just lets you quickly pop in to find something / browse everything with previews, metadata, etc. It works on Windows, Mac or Linux, works offline, has zero telemetry, deletes nothing, and is fully deterministic in operation. So, like, no goofy AI integrations for decision making or classification, everything is predictable and repeatable, you can even run portable versions off of a USB drive if you don't want to install it.
I also have a companion app RCHIVE Examiner [1], which is a free ncdu-like tool to see where the hell all your hard drive space went, and what's hogging it. Currently only in macOS App Store, but Windows and Linux versions are releasing soon.
There are a handful of discrete, decoupled components underlying both apps as well that people might find useful (especially in CLI scenarios) that I'm also planning on releasing gratis as soon as I get a chance to package them up standalone.
Check it out! I'd love any feedback anybody has.
[0] https://www.rchive.computer
[1] https://www.rchive.computer/examiner
I'm working on Zustri (https://zustri.com), booking and scheduling for therapists and salons. My girlfriend is a therapist and didn't like any of the booking/scheduling software she tried, so I built her one.
I finished the basics a long while ago. Since then I've added an AI agent, a public booking page, and support for multiple staff. Hopefully, I'm launching soon. Currently finishing final touches, gathering feedback and trying to figure out how to market this
Polishing up a paper on the minimal components, dynamics, and selection of any systems that persists. https://meta-r0ze.github.io/Persistence-Mechanics/Persistenc...
And then testing its completeness against the vacuum under the presumption that the universe persists.
A reimagination of bloodborne as a native VR game. All interactions have been redesigned to feel physical in VR, and combat retuned etc. It feels less like the original and more like a native VR experience now
https://github.com/bldbrne-vr/bldbrne-vr --- https://www.youtube.com/watch?v=doleBBoJL5U
My son is dyslexic. So I built a Chrome text-to-speech and reading tools extension to make it easier for him to get access to written content online (and also to support him at school). It has support for Google Docs and PDFs as well as the web. There are a few of these, but this one tries to stay out of the way, have better tracking, better voices and lower cost. Still more to do on the reading side - writing is next after that.
https://www.maylo.app/
Serverless Wordpress + MariaDB without requiring plugins or any modifications to customer code: https://www.agiler.io
The serverless MariaDB is important because it can scale to zero, so you have dev sites that cost less than $0.50/mo.
Also added MCP and CLI tooling that exposes all of the infrastructure + WP CLI + DB so agents can develop WP sites... which is nice because you can still use the WP admin interface, while using AI to develop the site, debug it, change pages and posts, etc.
Magic AI - I'm using it as a sanitized kids chatgpt for 7 activities: Story, image creation, chat, reflections, journal, learn and quizzes. Sends me daily summary on Whatsapp.
It started with building pi in the cloud - with pure cloud flare workers. I have it working now, and can spin up - the wrangler agent can fully run it as long as you pass it an inference key, including a file view, a chatgpt like experience, and ability to build and host custom apps. Can use alarms to wake up to do plain language tasks, etc.
Trying to figure out how to package this up, and where this would even be useful for other apps.
nothing special, nothing too important, just my hobby spomeniks map: https://github.com/mudroljub/spomeniks
https://tempmailboxes.com
Audion https://github.com/audion-lang/audion
I needed a code-first approach for making installations and interactive art. MaxMsp gets messy with very big projects. Python didn't really work well for music making and Tidal/Pi/Sclang are only meant for sound. So I needed something that has bpm synchronized threads, but can draw and do anything else like file IO, networking, hardware control etc. It works quite well now.
Aurora Glass, a living aurora animated across your Mac's menu bar.
It doubles as a status light for coding agents. But mostly I just like looking at it.
Free, no account, no permissions: https://aurora.radiantmachines.com
I keep saving blogs and articles but almost never revisit them. So I’m building an Android app designed to help me actually read what I save. You only get one saved article per day. Finish today’s read to unlock the next one. You can also save Twitter, Reddit, etc. posts separately and revisit them anytime.
Right now I’m mainly validating whether people would actually want something like this. If there’s enough interest, I’ll run a small closed beta for Android.
Join the waitlist for beta access: https://forms.gle/XWXmK8HUSmAtfms29
TO be honest, I dont think I will pay for this, though revenue can still come through Ads
Hunting bugs in the Python ecosystem using fuzzers and AI tools I develop. If you have a Python package or extension you'd like analyzed, just let me know.
https://github.com/ReviewToolkits
https://github.com/devdanzin/fusil
https://lwn.net/Articles/1067234/
https://buzzpot.app — a shared canvas for agents and projects, which I am evolving into a multiplayer environment for teams.
https://leaflo.app — a CBT journal that I am developing into a library of CBT techniques and mini-courses.
https://plainline.app — an AI presentation service with a focus on text and data, rather than visuals.
I'm working on a better way to design longer scrolling pages with multiple sections, while preserving the visual look/feel and drastically speeding up the experience.
Here's a demo: https://non.io/video/linen-demo.mp4
As a disclaimer I cut 32 seconds of generation time out of that video, but still I'm quite happy with the speed. Not launched yet, but it's been fun to work on / wanted to show off progress :)
Writing a Newspeak implementation of the PostgreSQL TCP protocol!. Talking to a GnuCash database from Chrome web browser. Uses my Newspeak TCP library. Both are bare minimum, probably very broken, but I am able to parse a Postgres result set from a simple query.
Here is a one minute video where I print the result data to the console from a Newspeak workspace.
https://youtu.be/D5wupSxi9Y8
I'm working on https://github.com/BowlingX/super-release, an alternative written in rust for semantic-release and mono-repo support. I build it mainly to make shipping packages in larger JavaScript/Typescript mono-repos easier. Also I started getting into rust more and wanted to learn the language.
Still changing along on Incredible (https://www.incredible.rs/) a Rust Text User Interface (TUI) Framework for the 2nd Quarter of the 21st Century.
It is New, Experimental, Modular, Declarative, Event Driven and Reactive. It builds cross-platform apps that can be compiled for the terminal (macOs, Windows, Linux), a standalone native window (macOS & Windows 11), or the browser (via WASM) from one unified code base.
Curious if this project will ever be 90% done.. ;)
HostsTray, tray utility to manage your hosts file, that also has a reverse HTTPS proxy builtin, that installs a self-signed CA and generates certs on-the-fly.
https://github.com/maxmart/hoststray
Also it forces your browser to "flush" their DNS by killing the keep-alive connections that for example Chrome keeps open in background. And it can be controlled by your local coding agent through a simple API (which also serves the skill for it)
I’m building Tinta, a journal app for Android where all the AI runs on the phone. It works in airplane mode once the models are downloaded.
You write or dictate an entry, and a small model (LFM2.5 1.2B) asks you a question to reflect further. It turns out 1.2B models are decent at asking questions, so I leaned into that. An embedding model finds older entries about the same thing and shows bits of them under the new one.
Most people using it are on budget phones running the models on the CPU, where the app provides good experience.
https://tinta.page
Great idea! I'd love this, but as an open source app. Out of interest (not a criticism), what made you decide to do this closed source, especially considering there doesn't seem to be a commercial interest involved (for now)?
Thanks for a good question! I see this app as an MVP that needs user feedback to deliver exceptional value in a world where cloud AI is abundant and doesn't use 2GB+ of RAM to run. I am excited to see what shape it will take, and it feels too early to determine source availability.
ColliePWA[1] a mobile interface to continue working with your agents when you're on the. Trying to meet devs where they work on the desktop in multiplexers (herdr/tmux/zellij). Remote access can be set up with any front door like tailscale.
Still not sure where this is going as the space is super crowded and moving fast, but it's solving a lot of my own problems and doesn't rely on third parties to run.
[1] https://colliepwa.dev
https://teamschatexporter.com/
Privacy friendly open source browser extension to export chats from web client of Microsoft Teams. Reached ~90000 users now. Currently working on making it more stable to export huge chats, which contains several years of data, while keeping things efficient and fast.
For the first time in along while, I'm not actually working on anything. I do some vibe coding for fun.
I'm kinda leaning towards vibe coding my own GUI coding agent. I currently use Pi and recently have been playing with Herdr and Paseo, but none of these feels right to me. These are amazing tools, but ai think we could do better. But I'm very hesitant - I know the only way to "succeed" in a project like this is to write (vibe code) hundreds of thousands of lines of code and it's kinda depressing.
I wanted to try and build something in a completely saturated market, so I’ve built a handful of movie themed daily games into a hub.
https://smashcut.games
Wow! This was really fun! Thank you.
Getting https://streamish.github.io/ ready to launch, a music server compatible with multiple 3rd party mobile apps and particularly Synology and QNAP. Want to do Plex next but I suspect that's a dead-end since I think you can't directly sign in you have to do it through their website.
Now putting it to personal use and keeping an eye out for issues and rough bits to smooth over.
Built with -
NestJS, Sequelize + SQLite, OpenAPI, and Jest
React, TanStack, Shadcn, and Playwright
I've built a habit of completing a game of sudoku every night before bed. After being frustrated with how messy my board can get, I decided to build an app for my reMarkable Paper Pro. In my opinion, it's a big step up from both paper and phone apps.
https://garbles.github.io/sudoku/
Local-first archive for AI conversations for providers without sane exports like Google AI Studio, OpenRouter (including chats), LM Studio, and others, as I couldn't find any chat archive with OpenRouter/Google AI Studio chat support. It supports semantic search, MCP integration, and continuing old conversations with llama.cpp. npx threadshelf https://github.com/ChrystianSchutz/ThreadShelf
Recovering from pushing out an article on coroutines [1]; coordinating getting the house in order for the workweek [2]; and drafting a package to make literate programming in org-mode a bit more to my liking.
---
[1] https://www.reindeereffect.com/0004
[2] Parts of the house smell of slowly cooking meal components (e.g., stock and beans), while others smell of cleaning products.
Over the last few months, I decided to learn more math and dive deeper into LLMs, RL, and robotics.
I ended up building a personal agent that asks me questions every day, emails me interesting reads in these areas, and helps me progress in my learning.
It started as a web app, but I have just ported it to a Mac desktop app.
It works like Conductor, but for your learning side quests: https://miletus.app
A little plugin to serve Marimo notebooks (which are great) inside an Astro static site (which are great): https://github.com/skritch/astro-marimo
"OneVote", which aims to answer: how much is your vote worth? https://github.com/skritch/onevote
A SAT solver in F#. The performance is surprisingly unintuitive, even for unsatisfiable problems.
Rebuilding my business’ website with Claude. It’s gonna be 10x better and migrate off Wordpress. My web designer broke something and stopped answering so giving it a shot and it’s going great so far. So stoked for the final vision I have.
I continue to update my coach app that generates workouts for me. It’s really solid at this point, better then the trainer I had, so mostly occasional bug fixes now or silly additions.
https://robheffern.palaich.net/sports/
I like sports! I like data! Combine!
This lets you see what the games on today are most similar to in the history of the sport. Click into one and see.
Also, the Sports Zodiac.
Like farmers of the olden years and the stars, we kinda set our clocks by sports these days. But what does that look like?
'Play the Years' to see a good timeline of how the years go by.
The war years, the lockouts, and the pandemic are just really easy to see in the data and quite stunning really
I'm wrapping up my 2+ years of work on a browser-based version of EverQuest that plays itself: https://idlequest.net/
Also have some other cool browser-based 3d game projects listed here: https://brynnbateman.com/
including an online version of Shenmue: https://newyokosuka.com/
Storyline isn't available yet, but about 75% done with it
A new kind of grid-based flowchart editor. The design is to have a grid instead of a canvas, and all nodes are the same size and sit in a cell. On top of that, there’s keyboard navigation kind of like a spreadsheet.
The idea is to make editing flowcharts way faster by being keyboard centric and having autolayout.
https://www.knotend.com
Have you started looking into different layout patterns? For kavla.dev (my SQL canvas) I'm exploring sugiyama (https://en.wikipedia.org/wiki/Layered_graph_drawing)
I like it, but I feel like there might be something I'm missing
Working on https://agentfriendlycode.com/
The project lets you find out which agent might perform better in your codebase. It scores the codebase for different agents using their static signals and gives you a response.
I'm putting the finishing touches on a new website for my LisaGUI project. The design of the new site is much more modern than the current site's design... I just have to finish rewriting the copy on one of the pages, and there are a couple of minor things to add... and then it will be done!
https://cool.computer/
A better, model-agnostic artifact experience to work with AI visually. You can annotate and draw on artifacts, so can your agents. You can share the artifacts with colleagues or friends and with your permission they can send comments to your agents too.
When you need more compute, artifacts turn themselves into microVMs automatically (paid feature) with a 8GB allowance
Currently we are working on Kumoss Open Source, intent-based infrastructure platform that turns natural-language requests into infrastructure changes, while keeping validation, compliance and existing engineering workflows in the loop.
https://github.com/InditexTech/kumoss
Basically allow developers to request any infrastructure need in a web portal and automatically that request is fulfilled following the organization standards the platform team defined.
Any feedback or comment is more than welcomed!
Any feedbac
I've been working on a language learning app.
At its most basic it keeps track of all your SRS-y things for you and builds stories in worlds you can create to help review them, but it's so much more. It essentially combines a bunch of things I've wanted as a language learner with my favorite aspects of various apps I use.
What languages?
Signal analysis to drive <desired signal-driven action>. Grouping/categorizing/stratifying signals into useful buckets to drive actions. Blind source separation has drawbacks like continuously guessing at sources and labeling them as brand new instead of having a memory of previously identified sources. Think of how humans can distinguish pianos from drums and remember the distinction across songs.
I have a service that crunches public filings in the financial sector: https://embarc.com
There is a lot of data for browsing, though it doesn’t aim to solve specific user problems as of yet (I did take a stab with RTOs in the logged in dashboard).
https://embarc.com/dashboard/companies/rto
Working on a Chrome (& Firefox) extension to make finding and downloading Free DLs on Soundcloud easier.
https://cratediggr.com/
Artists make users jump through hoops to like, comment, repost etc.. before giving them access to the download - CrateDiggr highlights the free downloads in the Soundcloud UI and clicks through the download gates for users.
Been a fun passion project, check it out if you download tracks from Soundcloud!
I've always been interested in health, longevity, and biohacking.
I'm Canadian but I moved to USA to work for Microsoft. One thing I liked about the USA was how easy it was to order bloodwork/lab testing. My original idea was to bring that experience to Canada, but since starting I've found other ways to improve the patient healthcare experience in Canada too.
My company is https://teletest.ca
Still cracking on with https://whistle-enterprise.com and making improvements to the document generation model. I'm ditching my old trusty Phi4 model and testing out some other small models to fine tune, something about Phi4 just seemed to work quite well for what it was so it's a hard contender to beat for midrange laptop CPU inference at a decent rate and quality.
A private managed cloud to host apps and run agents.
Give it a try: https://tarvis.io
Its like coolify but cooler since you don't need any technical knowledge to host apps. You can also run claude and codex remotely on the vps without keeping your computer or laptop open all the time.
I'm improving https://davegooden.com & just pushed https://edinacountryclub.com live as I continue to build out my real estate empire. (kidding, not kidding) Hopefully the industry can hang on for 10 more years.
https://www.haptrix.com - An Core Haptics editor, with the release of the iPhone Duo, the editor finally makes sense on device, so I've been working on the new UI.
https://www.phonique.com - An AVAudioEngine/PHASE Audio graph editor, it's easy to connect nodes together visually, then export the wired graph as Swift
I helped with Immersion Studio almost 30 years ago, very cool to see this. Don't have a Duo but will try it out.
https://archive.org/details/ifeel-immersion-touchware-deskto...
Wow, that looks wonderful, takes me back to my windows days, I love the logo too
I built Rindraw, a (hopefully better) Procreate alternative that runs on mac, ipad, iphone.
I believe that everyone is and will become an artist in the future, and everyone should be able to express their own unique ideas thru, especially in the age of AI
https://apps.apple.com/app/id6818370032
We just got a puppy, and needed a way to track potties to help us with house training. So, with AI/Claude's help, I found a similar open source package (Baby Buddy), forked it, and very quickly made it my own. Self-hosted, local for privacy reasons.
Repo: https://github.com/viciousnemesis/housetrained
Puppy: https://imgur.com/piDE0gq
I'm working on a TradingView Pine Script compatible Python framework: https://pynecore.org . And a Pine Script compiler for it + a full IDE for Pine Script with syntax highlight, language server, full featured debugger, charting, data downloads: https://pynesys.io . It is 99.771% bit exact on the top 1000 Pine Scripts and 100% trade match. I'm doing it solo.
I am building a tandem, reverse-trike halftrack, using the front end of an ATV and the back end of a snowmobile.
Replacing snowmobile skis with wheels is not a novel idea, but I wanted something which rides more like a motorcycle, with extra length for a passenger and a cargo basket.
It should end up looking like it came out of Mad Max; I plan to cruise around at festivals, running an ad-hoc taxi service, as a way of meeting people and striking up conversations.
A PR Review tool for ergonomics - https://bobby.viraj.computer
* files in organised trees/graphs generated using LSPs and symbol references instead of linear file list * keyboard shortcuts for everything from navigation, pan, zoom, comment, note, submit... * mostly github compatible with native features * LSP powered symbol resolution for inspecting definitions and references * LLM powered reviews supported * Inline LLM question answering
I've been building out a link blog idea called Linked List that I've wanted for a long time. Curation will be even more important in the future and I would love to have another option beyond Hacker News and Lobsters to find good reads suggested by fellow humans. It has been a fun space to explore ideas.
Not quite ready to open up sign-ups, but you can get an idea of the site here: https://linkedl.ist/@kale
I've started maintaining my own Firefox fork that:
- removes the stuff that I don't care about (Firefox sync, VPN, etc.)
- fixes a lot of the awful UX/UI choices that the (frankly) incompetent product team at Mozilla refuse to
- actually adds stuff that I do like integrated mail and calendar, a document library system (basically Notion), native ad-blocking, terminal, task-tracking like Trello, time-tracking like Harvest, etc.
- has integrated AI (Apple intelligence for MacOS for the easy stuff and Claude for the hard stuff
It means that instead of having half a dozen applications open, I can just work within my Firefox fork. Because it's all deeply integrated, I can do wild stuff like:
- tell my AI to search the web on a schedule and send me a custom daily brief with my own email
- maintain a comprehensive memory of all of the work that I do, across half a dozen clients, and be able to query it easily
- pull analytics from multiple web dashboards and format them into something cohesive
- program the web with a node-based infinite canvas that has live websites as part of the workflow
- manage my tasks across multiple domains (for example, it will see that I sent an email to a client and automatically update a task that says that I need to do that)
It's incredibly powerful and it has made me much, much more efficient.
I'm creating an IDE for founders to run their operations. Create automations (deterministic), dashboards, and soon notification systems. It's a desktop app, and the whole IDE runs in pair with claude/codex through local mcp.
It was developed as an internal tool at my other startup, Ipek helped us to run the whole thing smoother and with all the context I need.
https://ipek.run
https://sdkgarden.com/
been curious about the intersection between observability, SDKs, and product analytics. i couldn't find anything around SDK observability so built this little tool. curious what yall think!
i assume at bigger orgs there is some internal handrolled tool like this that ingests datadog/sentry/whatever info...if anyone plays around in this space and has feedback id love it!
I am building a system that connects texts (books, research papers, etc.) in a meaningful way to help people decide what to read and be aware of how texts relate to each other.
This sounds like a really cool project, do you have a repo or anything?
Not yet, but I am planning on making it open-source later.
A mobile car racing game with novel, one-handed controls: https://awais.io/simracer
I am building Abbox trying to help engineers focus longer on tasks without being interrupted by non-technical people with small requests.
I always got annoyed by change this h1 to h2, add tracking to share button, why can’t the customer play the video, the styles don’t match figma.
If you are engineer who also gets distracted I would love to hear your opinion and feedback. It is a bit hard to put it all together as every department has his own needs and opinion about the website.
My email kamran.elchuzade@gmail.com
I'm still working on https://openaltfinder.com, a directory of open source alternatives for proprietary software.
Trying to find out why Google has completely de-indexed the website, but other than "low quality content", I'm not getting any answers. Even Bing brings in more traffic...
Writing. I'm going to start an MFA next fall at LSU New Orleans. My latest piece, Three Black and Milds, a night in Juarez. No commentary, no lesson, just the city and the people in it, what it sells and what is bought.
https://www.metanoia-research.com/dispatch-007-three-black-a...
Just released this: https://github.com/rlorenzo/deadwood-rs
Since everyone is rewriting everything into Rust I created a tool akin to fallow (https://fallow.tools), but for Rust:
Finds dead code that LLMs leave behind.
Turning RSS, atom, blog feeds into audio feeds with kokoro tts. Free, web app only.
https://porchlit.app/
I built the original for myself because I wanted to listen to acoup and other history blogs while sleeping. And in the spirit of getting better at finishing projects decided make it public even tho it’s nothing really unique or special.
Still lots to polish especially in the text/copy and ui but the core features should be working.
I'm building an attribution and analytics tool for onchain apps. It's been a learning journey into data engineering and business intelligence best practices. Thinking of taking my learnings into a new product for a different segment!
https://formo.so
I've been working on an Obsidian plugin to quickly enter caloric and nutritional data (https://www.github.com/Krishna-Sivakumar/joule). The core features are done (A DSL and a runtime for finding and resolving formulae recursively), and I need to work on documentation and some convenience features. Hoping to finish it this month!
>Making a simple game
>Want player scripts to be possible (And dogfood it for enemy ai too)
>Behavior tree dep looks too easy, let's make a dsl + interpreter
>Typed builders are annoying (because stateful), lets separate into builders -> untyped ast -> typed ir
>Hm, typed ir is super repetitive, now I have a schema (ts) -> codegen (ts) -> .ts / .rs build step
Fun waste of 2 weeks, not even counting the time spent on pl literature
A domain search app called NameTheory: https://apps.apple.com/us/app/nametheory-domain-search/id680...
It uses Apple Intelligence to help brainstorm domain name ideas. I’m interested in seeing what small local models can do as they catch up to larger ones, this is my first iOS app so I started with something simple that scratches my own itch.
Maintaining my indie social network for tech people https://stackness.dev/ (currently struggling to keep users on the platform, even games and achievements didn't help, can you believe that? :D)
and this weekend I'll finally launch a tool for guitarists and their friends https://kantamigo.com/
Too many things...
I mostly focus on solving my own problems, and if I find a solution that can serve other people, I sometimes push it out as a product.
1. A watch app for runners that focuses on competing against previous runs, inspired by the rally games of the 90s where you could race your "ghosts"
2. A LEGO deal tracker that focuses on collectors and LEGO investors and is focused on price prediction based on data.
3. A mobile tracker for all the hundreds of components I use when tinkering with physical devices.
...
(Disclaimer: all of them are built using AI-assisted coding, and I wrote almost no code at all, although I have 30+ years of experience as a developer)
Working on a natural language intelligence layer for contacts - "Pull everyone I know who works at a Series A healthtech in the Bay Area who also went to Stanford for ungrad" => either structured or unstructured results. Large-scale data support, automatically enriched with up-to-date data, fast/accurate results. Integrates natively with Gmail and Gcal.
Looking to pilot for free with small qualified teams (early-stage startups with real/immediate need).
hello@llmsolutions.ai
I'm reverse engineering the streaming protocol for three different brands of cheap surveillance video cameras. Each brand has its own app. They're all awful. None of them have a native desktop MacOS app. I've successfully created a native app for MacOS and Android for one of the brands. They're faster and work better than the ones they're replacing. One down, two to go.
XMEye?
Would be interested in replacing, too.
A suite of own cloud based local first apps optimized for eink tablets, specifically the boox palma 2. Notes, book reader, browser, music player, maps, camera etc. With a companion ios app combining all of them.
this way when i want to disconnect i can just grab the palma and go
When I was a kid I loved a game called "Stars!" from 1995. I cloned the core and I'm also adding my own touch and lots of high sci-fi concepts. Already fun to play, early access coming soon: https://store.steampowered.com/app/5482440/Asteria_4X
I've been working on a voxel game engine called Bonsai for ~10 years.
Probably the most interesting thing about it at the moment is the editor. The world, and most things in it, are represented as collections of SDFs. More accurately, they're density fields, but, potato-tomato. One neat thing about representing the world as SDFs is they've been thoroughly researched and documented by a guy named Inigo Quilez, and they have a lot of nice mathematical properties. For example, you can do a smooth union of arbitrary SDFs to get nice rounded contours where shapes join.
The world generator and editor operate almost entirely on the GPU. The terrain and editor pipeline is a three stage process. The first stage is user-configurable; users write an arbitrary 'shaping' shader in GLSL. The second stage is a fused stage; it does a convolution on the output of the first stage to compute a normal, runs a second user-configurable 'decoration' function, and finally applies edits. Edits are collections of SDFs that are 'compiled' from a graph structure on the CPU into a bytecode, and interpreted in the shader. The last stage runs another convolution to compute the final normal, then packs an occupancy bit, color, and normal down into a 32-bit representation, suitable for transfer back to the CPU. I expect I will expand this to include additional user-level bits for encoding biome type and GPU-driven spawning.
The entire shader pipeline is one fused 'ubershader' .. everything is bound at startup and the shader is dispatched three times, incrementing a 'Stage' uniform on each dispatch. This strategy conceptually makes the driver overhead nearly zero, and the profile agrees.
The final step in the process is giving the CPU a chance to spawn additional entities and structures after generation. This is not completely fleshed out, but the bones are there. The idea is that the results of the GPU work are passed back to the CPU, which can run arbitrary callbacks to spawn additional edits and entities. Once spawning is complete, the result is meshed using a fast binary meshing algorithm and uploaded to the GPU. I keep 16 bits of data per voxel from the GPU work, 1 bit of occupancy, and 15 bits of normal for collision response, which works out to half a megabyte per chunk. At the moment I only keep chunks which contain a piece of the world surface, and discard the rest. Surface chunks are dense, and may be compressed in the future, but keeping this amount of data scales to world sizes greater than 1B voxels cubed, which seems good enough for now.
In terms of raw performance, GPUs are ridiculous. My 3080ti mobile gets somewhere on the order of 500B-1T perlin samples per second with a naive perlin implementation. Compared to FastNoise2, that's about 1000x more, which sounds wrong since my GPU only has 35x more capacity, but, that's what my initial profiling said. That number is the marketing number; the real throughput is significantly less due to multi-stage nature of the system, but, there's a lot of headroom for improvement. For example, dropping the backing textures from 32bit to 16bit roughly quadruples performance. Since we only really care about the surface where values are near zero, I believe using 8 or 16 bit is a viable option, though this claim will require a fair amount of study to validate.
All in all, I'm pleased with how this is all working out. My original goal was to write a highly flexible system giving users both easy-to-use tools and granular control over the final output. I think I've done a fairly good job at that, and the fact it's also super fast is the cherry on top.
I'll do some shameless self-promotion and leave some links here for anyone interested in looking at the engine, language code, or some pretty pictures.
https://github.com/scallyw4g/bonsai
https://github.com/scallyw4g/poof
That's so cool, building something consistently for 10 years is no small feat. How did AI helped you, or How did AI impacted your day to day work in recent years ?
Thank you :) One of the nice things about working on a game engine is it's actually a large collection of small projects disguised a big one. That let's my squirrel-brain jump from thing to thing and have enough variety to not get bored.
I just started using AI a few months ago. My smoke-test was "can it write a simple memory allocator". Until recently, every time I tried to get an AI to write an arena or simple buddy-style heap allocator, they just vomited all over themselves. To my surprise, the last time I tried, the AI absolutely crushed it; it unearthed tricks I didn't even know about in one shot. Since then, I've been using AIs to do the boring, rote work that effectively just boils down to me typing in something that's in my head. For that, they're great. I typically sketch out the thing I want it to make, then tell it to fill in the blanks. Sometimes they still go off the rails, but all in all it's a useful tool.
Where I've found they still struggle is making any architectural decisions, or where performance matters.. even a little bit. They often make super braindead decisions when it comes to performance for no discernable reason. They'll course correct when you tell them to, but I've had to keep a close eye on the output and reign it in frequently. Asking an AI to design a system from the API level hasn't ever returned a good result for me.. I always do that work.
All in all, I feel like using AI will probably make me a better programmer in the long run, contrary to popular belief. They accelerate the boring stuff I already know how to do, which leaves me with more time to work on the edge of my skill level where I grow as an engineer. Guess we'll see in 5 years.
A video game in which players can build their own worlds. The brains of the creatures in the video game are a small language model + harness that runs in your browser: http://id-le.com
It’s open source: https://github.com/nicobrenner/id-le
I'm still working on my own shell based AI tool. This month I am testing and reworking it to try and get useful results with really small models, like 2B level. Goal is to get help debugging Linux issues when offline with these small models that can run anywhere. Then also being able to access cloud based models to do harder stuff once online. all while feeling like a CLI tool.
github.com/patrickjh/ssa
We've been reverse engineering the tensor core on TPU v5: https://gherkinlabs.com/v5e-tensorcore-instructions
The work mainly stems from a belief that Nvidia gpus' architecture are a bit subpar for machine learning and that tpu could be a viable alternative but are still rather undocumented.
Ok, I assume google would have an API documentation. Why would you need to interface at the instruction level?
I've been building data pipelines from sometime as part of some other products, eventually pulled out a part of that system into a standalone product called hydrafetch.com
That's the main thing that I'm working on these days
Working on a tool to provide a thin OIDC auth layer for self-hosted LLM engines. Users log in with your Org's identity provider, generate an API key for themselves, point their harnesses at your instance of the relay, and then they're off to the races.
No web chat interface, no call-us-for-pricing feature gating, just a tight Erlang program for people who want to get shit done!
I’ve been working on my inertial navigation library. It actually beats ArduPilot’s navigation subsystem in my benchmarks. I’ve been working on these algorithms since my PhD thesis and now I am releasing the stuff as open source embedded C and Python library:
https://github.com/jnz/inslib
I'm building mensara, multi player memory so you're whole team can work together with ai coding software.
One of the most fun things is you can use all the various agents, like muse, grokbot, dots etc and they all know you immediately and can work together.
Try it at www.mensara.ai
A cli command to view PDFs, images, Word documents, spreadsheets, drawings, Markdown and HTML inside the terminal (kitty, ghostty, etc.)
Works well with tmux and through ssh too:
https://github.com/dobladez/clview
I've been working on a small search engine for content creators that play and/or review indie games. It's written in C and has been a lot of fun to work on.
https://marketyour.games
https://prophitt.me/writing/a-modern-web-app-in-c
Still building the file manager I've always wanted: https://intestinate.com/gusco/
Added an audio player this past month for all fellow file hoarders out there.
Also, you can now choose if you want Gusco to live in your terminal or run as a native-yet-texty app.
Was pleasantly surprised how well turning a Bubble Tea TUI app into a GUI-ish bundle works with Gio UI.
Vicena.ai allows you to design, simulate and test new battery chemistries, new drugs… With the same ease as building software with coding agents.
It is an AI for Science connected to scientific compute and a network of labs, so you can do science end-to-end from your computer.
https://vicena.ai
Infrastructure for hosting static webpages with options for storing data, having users, sending notifications, shared "rooms", and various other bits and bobs all behind an mCP. All $0 practically because it's made up of static services from AWS.
Made for me, but I can provide access to friends and family, which is nice.
My integrated terminal + file browser with enhanced autocomplete and search. Replaces all native OS ones for me: https://github.com/David-OConnor/peregrine . Also includes misc utility OS functionality, tailored. (Advanced clipboard, a scratchpad etc)
I started a company to focus on the niche accounting professional service called Cost Segregation in the summer of 2024. I was laid off from Meta in February 2025 and dove in full-time on that startup, Segtax.
We are on the cusp of automating the bulk of the work. It's been an exciting ride
What was your work at Meta if I may ask?
I just celebrated the 1 year anniversary of Pixel Brite, the media server Plex never became.
https://demo.pixelbrite.site
What makes this different to other media servers?
I’ve put considerable effort into answering that on the site @ https://pixelbrite.site/en/media-server
Proper HDR and ABR are personal favorites. Performance and efficiency (memory, cpu, gpu) numbers are posted there for comparison.
There are several interactive demos available on the site worth checking out
The alternative is asking users to buy $749 Plex Pass or settle for SDR only Jellyfin. I wouldn’t even want that product for myself
A modern take on the game FTL: Faster Than Light, with inventory autobattler gameplay. It is called Cosmo Cargo and a demo is available on Steam: https://store.steampowered.com/app/4659480/Cosmo_Cargo_Demo/
I've just released v1.0 of Rouler which is VPN split tunneling for windows. Think IP rules for Linux, but with per-app/service connection steering. The whole app is built in Rust, including the kernel driver. Having released v1 I'm now starting work on adding folder and windows store support. https://rouler.io
https://petrosprawl.fyi/ An attempt to map oil infrastructure around the world
https://chord-network.vercel.app/ A tool that helps you build chord progressions
Absolutely loved petrosprawl .. bookmarked. Do you provide structures data as well for same?
Ive been blogging about learning how LLM services work. Just finished the 2nd part and hoping to do the last part this week. https://matthewtejo.substack.com/p/designing-an-llm-chat-ser...
Building an intelligent typing application. https://typequicker.com
Anyone can learn to type +100wpm - just have to focus on your weak points and spend 5-10 minutes a day for a few months.
https://github.com/Polign/recall
Polign Recall, a state and memory management for agents that can provide append only state logging and governance. The memory and storage is completely on BYOC setup and can work with confidential workloads
I’ve been working on an accounting agent factory (https://www.translucent.io) . Building a harness and infra that can process books with thousands of invoices, transactions, etc. is quite rewarding and challenging. In particular the long-running nature of things.
I recently made a mobile app for practicing singing: https://singtrace.karamagic.com
The free version you can record and keep an archive of your daily practices of each song, while a paid version does some analysis of your vocals.
Still working on ShopSpec, a parametric deterministic furniture builder. Trying to find time to put more actual builds in the gallery - have a few projects that are in progress while continuing to build out the tool and iterate where I find personally useful.
https://shopspec.io
A free comedy newsletter that has nothing to do with technology yet mentions it from time to time.
https://substack.com/@yelluwcomedy
Started making youtube videos again, using my human brain to pick the topics and write the scripts, and AI to edit them (mostly).
Here's one about speculative decoding (and how it speeds up AI inference): https://www.youtube.com/watch?v=cGnIKjGIwjA
I have spent the last six months or so on a multiplayer, open world game. The idea is you can jump to any city, walk around, see businesses and chat with others. Loading the resources (map etc.) as they are needed and garbage collect have been a challenge,
Building a soccer game based on Mike Tyson's Punch Out, just to see if I can create some method to see if an agent can iterate on making something more fun. Obviously that depends on the definition of fun (which is already a can of worms), but it's a fun way to use a different side of the brain.
Answering the second question:
I've been curious about...
The machinations of my own mind/psyche/nervous system. For various reasons, I have had to get increasingly curious and stay that way over a years-long haul. This includes the usual suspects:
- "Serious Stuff", lets say. Viz. paying attention to and learning about therapeutic introspection, relationship/social dynamics, clinical psychology, meds as they apply to me.
- BUT also can include "fun stuff"! For example: seeking out and befriending like-oriented kindred spirits (it feels great to be seen / heard by peers), and developing a persistent interest in biology (which I hated in school), and even signing up for research studies! I got accepted into a Phd memory study and it was great to be back on a university campus, a couple of decades since graduating.
Current opinion: I think all mental health work leans N(1).
The art of it appears to be remaining curious and seeking self-understanding without lapsing into tiktok pop-psych, nor a false sense of studied erudition (No; one will almost certainly not become a clinician).
The greater the complexity/depth/intensity of one's condition, the more it calls for subtle, bespoke work. Progress from outside-self intervention has an upper limit (e.g. "taking the prescribed meds"). Active, self-directed self-work (in concert with one's support system, of course), brutal as it, must be done... somehow.
To others in a similar state space; may The Force be with you. Following one's curiosity, even now, will probably not result in the curious cat's fate. Worth the extra (omg so much extra) trouble, IMHO.
_|/
P.S. Anyone wishing to discuss more, feel free to email me (see bio). Please refer to this exact comment.
I've deliberately not shared specifics, including self-educational resources, because this stuff is slow, plodding, deliberate N(1) work, and I've personally found it counterproductive to read/consume the vast majority of N(1) experiences I've found.
(edit: fix typos, reformat for clarity)
For the month of November (and starting in October this year) I like to hold weekly cooking challenges to practice and refine my Thanksgiving day feast. Now about 3 weeks out from the final boss level of cooking, and I'm having a week all about brussel sprouts. Earlier weeks were cauliflower and the classic seasoning blend of sage, rosemary and thyme.
https://hn.fiodorov.es AI-powered search across all HackerNews comments
https://climate.fiodorov.es Agentic analysis of climates of world cities with more than 300k inhabitants
Both have mcp.
A few projects, but now waiting for a week for my Piano Game[0] to be approved on Play Store. Maybe it takes longer cause I marked it as being "for kids" too.
[0]: https://ultimidi.com
I'm working on a data monitoring company:
https://turbolytics.io/blog/debezium-wal-growth-postgres
Happy to get your postgres and kafka connect monitored and alerted on for free!
danny@turbolytics.io
I've been updating the landing page for my book https://networksbyanalogy.com website to be dynamic so users can "explore" the book in a visual way.
Sysadmin agents on old machines.
Run AI on machines 30 years old.
Entirely in legacy Perl.
HP-UX, OpenStep, Solaris 8, Tru64, Irix, old Linux. Legacy FreeBSD, NetBSD...
Optional desktop in tk.
There's even a proxy for modern tls.
https://github.com/kristopolous/old-agent
I've been working on https://gitsocial.org/ for a year and a half now. The approach is to keep it all in git: issues, PRs, comments, releases, etc.
Hoping it becomes a viable self-hosted alternative to GitHub, GitLab, and Codeberg!
I have been working on a derivation step auditor for checking any mistakes in your (or your AI buddy's) mathematical derivations!
https://samosa-math.vercel.app/
An Obsidian plugin that turns handwritten reMarkable notes into searchable text. You can run the handwriting recognition locally on a Mac. The latest feature I added sends your highlights and margin notes from the tablet to Readwise. https://taggedsync.com
Building a native macos app with a suite of completely local AI tools. The first version is out and have a long roadmap in front of me. Would love to hear some feedback.
https://artifactfour.com/
Tritium, the legal IDE: https://tritium.legal/preview
Building the web-embeddable legal IDE for transactional AI-native law firms. Docx, PDF and Markdown in the browser with redlining, go-to definition and more. In Rust, of course.
Launched a word logic puzzle game called Meshword: https://meshword.org
It combines nonograms with crosswords, requiring both logic and word knowledge to solve. I am still working on the interface and tutorial, as well as the puzzle difficulty levels.
I built Telemetry Machine (https://telemetrymachine.com).
The TLDR is that if you like Honeycomb or Datadog but don't want to pay insane prices, you should check it out.
The longer story is that I love talking to SREs, platform engineers, etc about observability, so if you want to kvetch about it, please send me an email at jack@telemetrymachine.com!
My cofounder and I have been working on observability for a combined 15 years and we decided to take what we love most about all these tools. We've got the powerful querying of Honeycomb, the great UI of Datadog (though most users use us via agent anyway), and the speed of Clickhouse.
Check us out!
I'm building Oculette (https://prelaunch.oculette.ai/).
Basically, I hooked up an ESP32 camera to a vision model so I can prompt it to watch stuff and send webhooks. Thinking analog gauges, legacy equipment, sourdough starters (still no API), and stuff like that.
Been dogfooding the prototype on my desk, and now I have an air quality API that reads the display on a cheap monitor.
Also added direct camera access via MCP so Claude could help me debug the hardware, which has been pretty fun.
cool stuff. I have cameras watching some animals and I've been working on real-time analytics for their behavior.
It looks like this is sending video signal off into the cloud? Are you sure people will be OK with this? Can I integrate it with local inference and lock it down?
Why can't this just run on a spare cellphone? Why the dedicated hardware?
I'd be curious to see some technical details on how you're routing b/w the user's prompt to the continuous monitoring. Are you purpose-building a tiny CV model for each prompt? Or are you running a big model continuously.
Reflect [1], it’s a local-first privacy focused self tracking and data analysis app where you can set goals and run self experiments
[1] https://apps.apple.com/us/app/reflect-track-anything/id64638...
I took on a software project for a factory. During work hours, I write code for the company, and after hours, I work on code for other companies. I also submitted some PRs to open-source projects this month.
A fully hand-written project: https://github.com/smy20011/onecut
Edit video by editing code.
Building a replacement for docker focused on better isolation and more lightweight resource utilization.
It's a sandbox that you can embed ala the sqlite of VM's.
Link: https://github.com/smol-machines/smolvm
https://www.nichess.org - chess with health points
Working on a modern implementation of a torrent client [1], it's already pretty usable for most scenarios. Provides both a daemon + web UI out of the box as well as an embeddable engine for any Rust application.
[1] - https://spill.rs
I'm trying to make a video to Lottie Animation Converter but so far it doesn't work: https://gen2d.com/animation-video/
Software development is dying. I am teaching myself how LLMs really really work with this tool that I am building https://modelmathics.com you can try it on https://try.modelmathics.com
Still working on my audio FPAA ASIC for Wafer Space run 3!
Hopefully someone will soon have tested chips from the 2nd run that use 2 edges only on the pad ring. This would enable significant area savings.
The deadline is fast approaching and many blocks are only partially validated...
A south-park style animation creator, where you paste in a script and it will animate a short episode for you with voice too!
https://limittox.github.io/south-park-animator/
No example(s)?
Building my an airflow pipeline focused on collecting news.
Partly because I want one, but also because it makes for a good learning platform. e.g. initially do some of the processing with a stronger llm...save result and then attempt to train a SLM or similar
I am building Cloud based context compiler and intelligent codebase visualizer for Developers and AI Agents, for easy PR review, onboarding and easy architectural understanding
https://github.com/devlensio/devlensOSS
AI client for data analysis.
You can connect Claude desktop to databases using MCP servers, but Claude desktop is not great at displaying large volumes of data. A custom client can do a much better job.
https://reportr.com/
That's a crowded space right now! We've been exploring augmented analytics for a few months at work
What does it do differently to https://getnao.io/ ?
It is a full-featured reporting tool, and can be used without AI. If you do use AI it has a semantic layer to help AI understand your table and column names, and relationships between tables.
Trying to get the kids outside: https://eyesoutside.org
and letting families hold off on screens for a few more years: https://readychime.com
https://docking.cc is a Linux dock bar with Wayland, X11, Linux, Free/Open/NetBSD and experimenting on Windows
It has 60+ native applets like WhatsApp, Hacker News, Reddit, Calendar, Weather, ...
I'm vibing an old style RTS game and exploring how far can I get with local models for drawing, 3D models, rigging and sprites. I love how much I can accomplish with Claude in terms of code, but gluing it all together with graphics still requires plenty of "manual" work.
Werkbook (https://werkbook.app), digital forms and photo reports for field crews. I'm a solo Rails dev, and thanks to Claude it now has native iOS and Android apps instead of Hotwire Native web-view wrappers.
https://vester.si/motion/?e=plasma
My new favourite shader running realtime over your camera input. Has gotten great feedback from kids as well :)
Knowers of plasma pong might see slight resemblance.
I am developing the concept of a "digital environment of emergence."
Ur is an (very in-progress) universal formatter and linter for every programming language. It's in preview right now, which just means friends are messing around with it and having a good time pointing out all the bugs. Docs are mostly up to date, but buyer beware. Implementation wise, it's the Rascal meta language reimagined in Rust with some features from JastAdd thrown in there for good measure. Ur is incremental/persistent between runs, it's built on top of Mambo, a library that drops in as a replacement for Salsa and persists all queries to disk automatically.
https://ur.dev/en/
Ur has some pretty good static analysis and code rewriting tools hidden inside of it that'll get documented later on. These tools have been developed in tandem with a static mass rewrite of some core parts of the Rust compiler. It's very much not done, but it's made pretty good progress towards putting all crates in one process, dropping memory after a crate is done compiling, and full incremental support across all commands. Here is a literate program that explains why, how and which changes are being made (I haven't read all of it myself yet, lots still changing). Aiming to have this out by end of the month.
https://breakneck.powderworks.dev/
https://trylendify.com/
I'm working on an AI powered loan servicing platform. We've been trying to get cold / warm intros to onboard a set of design partners
Been working on my own Orca-like meta harness: getlattis.ai
Use it every day, and my team mates have started using it and loving it. In my dreams it’s a business. In reality the feedback I get from people that like it absolutely makes my day.
https://geotastic.us/ An AI analyst for ArcGIS. A lot of city data in the US is stored in ArcGIS or Socrata. This tool uses LLMs to generate code to analyze it.
Line drawing and maps, oh and global illumination and 3d printers... fun times!
https://blocks.fabiopiparo.com/
https://chattymaps.com/
An easy way to find a place that works for everyone. Think of it as a doodle for places.
You can also create collections of places, ask for recommendations etc.
https://github.com/ivankovic/stop-bots
Trying to keep my selfhosted server safe with a Rust TUI for managing bot blocking at the nftables and nginx level.
AGPL v3
I’m based in Berlin and currently working on ChatFrom.io . You upload your knowledge base, and it uses advanced RAG with hybrid search and reranking to find answers when someone asks a question. It connects to Stripe, Shopify, and other tools, and works on Telegram, Slack, and more. Free to start.
domain is showing avaialble for sale?
For the last year I’ve been building personalized tools to help my wife with her day trading and have started to productize them as https://cuppa.markets/
An MCP server to act as lore persistent record and tools to orient an open world text adventure game played through AI agents.
The theme is an empty Earth. All natural Earth with no human constructions (until the players start playing and occupying it).
Using Rust and RocksDB.
eremoterra.com
Drive.py Manage files across your connected and disconnected drives
https://github.com/web3cryptowallet/drive-py
A desktop app to easily browse, tag, and rate your images. It doesn't touch your files, it just reads them. Everything is stored in a local SQLite database. Think Eagle app but private, offline, and easy to use.
https://hatamoto.dev - a complete replacement to Wordpress and its most used plugins built into the system. One go binary, no node , no compilation, Postgres and sandboxes plugins. Built in AI analytics
I'm working on judt.ai
Trying to make an agent first too for agents to ship full micro apps.
Built on the awesome celld.dev which gives each app full execution environment, db, durable objects, etc..
Tracking the spread of pirate scene releases, but I can't say more publicly. And playing Old-School RuneScape.
https://www.occasionful.com/ an unofficial Partiful client that is easier to use than the official website / app
https://getmoulin.com/
An audiobook player that plays local audiobooks & librivox books. Also has a curated library of audio drama podcasts.
I'm working on a platform for selling event tickets (kind of an alternative to Ticketmaster or weezevent). I'm working on it on my free time for now (I am also using it to test working with AI tools ). Not sure yet if I want to build a company or not.
Also working on AI music video, trying to reach pro level, here is my last creation: https://youtu.be/aC4WQyKc_A0
The 2.x series of NoFlo
https://bergie.iki.fi/blog/noflo-20-prep/
neat!
I've been building MiserablyUnemployed for the better part of this year. https://miserablyunemployed.com/
It's a free job site with some resume tools bolted on. Lots of people from all over the world are using it and that feels great.
It has 2.7M+ jobs and automatically removes any job older than 7 days. We never allow reposts because, frankly, the job market is a nightmare and any job that has been up for 7+ days almost certainly has 1000+ applicants already. You're just feeding your resume into a wood chipper.
Come and try it. It's awesome.
Signed up and it is in fact, very awesome. I think it would be helpful to be able to select multiple regions, particularly for Asia / APAC
That's a good idea. Will add :)
Aficionado (https://aficiona.do) - a place to discover, scrapbook and remember media (movies, video games, tv shows, music and books) that resonated with you or that you'd like to try.
developing a decision-management system for code (https://edawise.com/deccy-framework)
and on the side, trying to develop ai interfaces that aren't built around conversations, currently in the form of a restaurant search tool but i can't get it fast enough (at a low cost) to feel usable
working on AudioWeave, an ad free, offline first Android podcast player.
switching from iOS, I wasn't happy with the existing podcast apps, so I built this mostly for myself. I've been using it for the last six months and finally pushed a public beta.
My day to day is BE so I used the project to learn Kotlin and Jetpack Compose. Everything is strictly native using Room and Media3/ExoPlayer. I wanted a nice looking (subjective), privacy-focused app without any tracking or ads.
I'm considering open sourcing it soon.
https://play.google.com/store/apps/details?id=com.enyodigita...
Indieweb/blogosphere index, search, and exploration.
Taking a break from features to focus on refining content filters and expunging sites that slipped through. Looking at you, .ai TLD.
https://outerweb.org
Unity for photos.
A tool that builds an FPS video game(1) in one day.
https://www.free-visit.net
(1) : The FPS game runs on PC, Mac and even Smartphones.
fxdax - https://github.com/semistrict/fxdax a port of fuschia fxfs to linux on top of a novel virtio device that stores file data in object storage. Intended use is file system for VMs that supports dax to avoid page cache duplication while maintaining high performance
https://khumbu.im - app to study german language (extract highlighted words from paper, generate anki cards etc)
https://github.com/a-poor/ganymede
I'm working on a native Jupyter client, written in Rust!
https://schmolt.itch.io/wormlove
A little puzzle game where you make worms kiss. 32 levels, plays in your browser.
Building out an interactive fiction platform which blends AI storytelling, images, video, and TTS with user choices at the end of each page in the story. Https://cyberstory.ai/
It started maybe 9 months ago with the idea of an AI driven Oregon Trail adventure and i've carried on with more historical fiction adventures across time plus alternate histories, romance, and more. Users can also build out their own stories.
One strange thing I ran into was the LLMs always naming characters the same names across stories (and i'm not using seeds!).
I've been doing a bunch of differential analysis and trying to put some of my large amount of private unpolished work online. Found a bunch of bugs in PNG libraries, Game Gear emulators, SDCC, and other things. I've been running a bunch of test ROMs on multiple revisions of hardware, and oh boy is that thing poorly-understood, if emulator errors are anything to go by. SDCC is the compiler I'm using for the ROMs and it's full of miscompiles and performance issues, so I'm running a patched build with everything I know about fixed, because I kept running into stupid issues while making the aforementioned ROMS (downloaded the whooole issue tracker and even fuzzed it a bit). Also running a patched build of Mesa after running into several bugs when trying out mesh shaders, a patched Fossil (https://fossil-scm.org/) because of bugs I randomly hit, a custom printer driver because the stock one sucks, custom firmware on my Siser Juliet cutter/plotter machine, ... and of course, barely any of this is anywhere online, and at least half of it isn't anything I wanted to do in the first place.
Anyway, I haven't shared this anywhere because I don't like how much AI-written wording it has, and I haven't really done any of the self-written analysis I wanted to put in there, but you can see the PNG analysis results at https://wizardmaxx.ing/difftest-png/. It compares against the 3rd edition, and you can select the 2nd edition if you want to for some reason. It's still missing an option to show what you'd care about by real-world judgement, but ¯\_(ツ)_/¯
I keep doing the independent research in theory of AI despite my odds
I was recently laid off, so I've been pouring myself into a new paas offering and communications suite. Hoping to have something to show off here very soon. We will see.
Minimal recipe manager
https://prepbook.app
i've been playing with this silly "game" inspired by one million checkboxes except the goal is to sort the board. something to kill time https://human-sort.app i like to think it's the same sort of work that's being done on severance
Lately my super fast anti gravity racing game. If you enjoyed F-Zero or Wipeout, you might like it.
https://nitelore.app
A bunch of small apps, kinda getting SaaS out of my system.
https://radziszewski.software/apps
https://github.com/hnarayanan/hello-guix - A GNU Guix installer for Apple Silicon Macs. :)
https://www.lungy.app
Interactive breathing exercises. Working towards a medical device version for asthma and COPD.
Working on Notekpr a local first and end to end encrypted alternative to notion and obsidian.
site: https://notekpr.md
Obsidian is end-to-end encrypted: https://obsidian.md/help/sync/security
I've been posting stuff on my blog https://blog.willmeye.rs
https://bloom.site
A place for independent film teams to capture visual references.
I vibe-made a Russian life simulator called Frontline Rush:
https://rawdod.com
I vibe-proofed the ABC conjecture (IUT LEAN Fable):
https://github.com/oneman/motizukipdf and https://1234567890.help/
Yeah
The official website has launched for MinusPod
Remove ads from Podcasts and YouTube
Website: MinusPod.app Docs: docs.minuspod.app Demo: demo.minuspod.app
Just finished a little animated short story: Washed Up.
https://mythitos.com
My main project is trying not to lose my mind, given all the things I can now create thanks to AI.
Jokes aside (although I'm not entirely joking about my mental health), my main project is https://auradistill.com: First-principles thinking for your AI agents.
It's not just about giving your personal AI agents "memory". It's about encoding memories in a way that can handle ambiguity and contradictions.
That project alone has got me completely hooked on building for and with LLMs.
I've got other projects too, but this is probably the one I've poured the most time and energy into.
I’ve been working on a web application framework in Go called https://atlas9.dev
forking the internet. taking internet technology and using if for community rather than (pick 1: greed, corporations, predation).
I'm a really big fan of Grand Theft Auto IV but its online scene never took off. I'm trying to build a multiplayer client a la FiveM/RageMP for it.
Triplec.fm
A not for profit, with the goal of helping more people listen to, and go experience live music, in Canada.
I'm working on a relational database called Turborel and a log-structured storage engine I maintain TidesDB.
https://tradeinsight.info
Goose-based internal Matrix chatbot / custom agent to automate some tedium away in my company of ~1.
https://arcaide.foo - to visualize the architecture of your (LLM's) code.
Adding tufting support to RapidCAM.app for CNC run makers. RapidCAM already supports 2.5D CNC, laser and rotary axis so tufting was a good fit.
small OSS projects to explore AI futures + safety:
https://doom-or-bloom.com and https://cultural-alignment.com
StatsBrowser: https://www.statsbrowser.com
A statistical analysis suite that runs in the browser via webR. It provides step-by-step guides for running tests and insights on results containing statistical pitfalls. Aimed at students who struggle with traditional stats apps.
working on talk deep https://talkdeep.me, a little app for keeping in touch with friends after i moved countries. you pair with a friend and get one question a day, or go through decks of deeper questions together. i've never done an app before so building something from scratch has been a really fun creative outlet
been building BuildOpsy mostly writing about system design, infrastructure failures and the occasional “well, that was an expensive outage” postmortem.
Also building calculators for things engineers usually calculate in a spreadsheet at 2am. you can see some on the site(https://buildopsy.com/calculators.html)
Still Raygum! Letterboxd for music - https://raygum.com
Working on https://vostride.com/
Not really a "thing" but I'm trying to come up with effective strategies for LLMs to generate screenplays that are not terrible. I think I have some that seem to work.
Reddit alternative https://hissy.app it is currently offline though
working on an app which you/agents can put exact pieces from any websites you need into a one screen.(https://banquish.space)
I didn't want to switch/refresh bunch of tabs. That's why I made this one.
not screenshots. so you can still interact with the pieces like in normal web browsers.
I want to make app like action but for repairs, like electronics and cars.
Client posts request and masters respond with their offers.
What you think about that guys?
There is already a site here where you say what your job is and garages submit their offers. They can also auto-submit offers if it’s something routine like a standard service.
Hey, Thanks for info. I am from Central Asia, could share a link for that site?
I feel like this is the third one of the November thread!
That said I will say again, working on http://leafle.org my app (mostly for me) to find new books to read that I wouldn’t have thought of (mostly recently Cosmicomics) based on a deeper understanding of what I like in books than people who read X also read Y, that I can actually borrow right away from my local library. Works on ~6,500 public libraries around the world.
Fun project to play with reverse engineering using LLMs, and jev and things like that.
I’ve launched two projects:
1. Novalede - https://www.novalede.com
I got tired of having newsletters in my inbox that I never read, and emails building up where you have to hunt for the important part, and not really being able to get the gist of whether the email affects me or not.
So I build an MCP connected newspaper. Ask your agent to publish them as articles where the lead is front and centre, and I can read it first thing in the morning. It’s surprisingly effective and catches so much context I’ve sort of stopped reading my email directly.
I also get it to pick the HN articles of interest to me and give me a summary of what the comments say. It’s nice to remove the sensationalism of how people react.
2. Bebril - https://www.bebril.com
This project is a fun one where you to state who you want to become and you are given mini quests to take that help you on your path to get there. I want it to be a choose your own adventure where you feel rewarded for action.
i've been building wearover and getalcor.com -- a chain of apps handing off content where possible with use of on-device AI , but burning tokens and building more than serving the product well to end users, and might just wind up my startup experiment and take up a job or contract.
I've really been enjoying using Codex as a headless remote agent on my home lab. I'll often pull it up from my phone and jam on some little project when inspiration strikes.
I kept bumping into the question of how to give it access to the things it needs without leaving my private keys and other credentials on the server though.
For SSH (and sudo through pam_ssh_agent_auth), I ended up building agent-auth and Agent Witness. Codex sees a normal SSH_AUTH_SOCK and supplies some context about why it needs a signature. If one of my Macs is available, agent-auth proxies the request to the 1Password SSH agent there. Otherwise it routes the request to Agent Witness on my phone.
Agent Witness runs as a PWA and is essentially a remote ssh-agent. The private keys are wrapped, encrypted, and stored in IndexedDB locally on the phone. The vault can only be unwrapped with a secret derived from a passkey, and all key handling and signing happens locally through the browser's Web Crypto APIs. The unwrapped keys are non-extractable CryptoKey objects (the web crypto APIs are pretty awesome!)
When a signing request comes in, the PWA gets a push notification. It shows me the reason Codex supplied, the command being run, and the key being requested. I authorize it with my passkey, the PWA signs the request, and only the resulting signature goes back to the server. The private key never leaves my phone.
https://github.com/evanpurkhiser/agent-auth https://github.com/evanpurkhiser/agent-witness
I also recently put together psst, which solves the related problem of handing a secret to Codex from 1Password. I give it a 1Password item-sharing link, it opens a local session where it decrypts the 1password item in memory, and Codex can inspect the available fields of the item before explicitly retrieving only the ones it needs. The CLI refuses to write secret values to a terminal, which helps keep them out of argv, shell history, and transcripts.
I’m sure I could work out a better approach where the agent absolutely cannot retrieve the secret since if it really wanted to read the secret, it could of course pipe the value into cat or whatever. The goal is to just make secret access explicit and avoid accidentally spraying credentials into places where they'll be retained. So far it’s been a very nice flow with agent browser.
https://github.com/evanpurkhiser/psst
tirreno - security framework
https://github.com/tirrenotechnologies/tirreno
Live demo: https://play.tirreno.com
Automating so many terrible chores with AI. Recently I automated pulling transactions from my banks (Canada, no open banking..) and putting them into Actual Budget. SimpleFIN or Plaid type services frequently break, make mistakes or require reauth every few hours for the banks here which makes them not so useful. Now I'm automating computing my taxes fully which gets pretty complicated because of international income, different assets etc. Its basically a data lake to process all kinds of financial data in various formats and compute stuff, moving lots of manual drudgery and complicated excel sheets into code.
I am building Esara – the AI hiring platform for candidate sourcing and explainable evaluation.
We’ve just incorporated in Berlin together with two great co-founding colleagues. Honestly never felt more excited.
Though it is a B2B tool, check it out: https://esara.ai
Trying to grow tomatos through winter.
My next sci-fi or fantasy novel - it could go either way at the moment.
Why not both?
That is what it is at the moment, but my experience from crossing genres in the past was not great so I will probably lean more heavily on one for the final book.
Koaches.app helping prepare and practice behaviorial interviews.
My husband created an AI-powered website, and I've been using it to generate engaging AI videos for social media. This not only helps promote his website (seedeo.ai) but also expands our website's content library, which gives me a great sense of accomplishment.
Finding purpose in life.
I am working on a Rust-based ext4 driver. The idea is that it can be re-used as a FUSE-based driver as well as (Windows) kernel driver (eventually).
Most of the essential features are implemented so far, but some (mandatory) features are lacking (xattrs support). My goal is to get to a full ext4 implementation that also supports ext2 and ext3 read/write.
The initial motivation was to have a pet project with a use-case for Rust, as I wanted to improve my Rust coding skills, after I did some work on a game server implementation in Rust earlier [1]. My goal is to have an implementation where I have digested every line mentally, which is something that pretty much excludes slop-like use of AI for large scale code generation. And of course I want to really trust the code that touches my files via raw hard disk access. Hence, I use AI mainly for unit tests and (scoped) code rewriting when my code gets too verbose and cumbersome.
In the long run I'd want to add fuzzing to the implementation, so if anyone has experience on this, I'd be interested in getting some pointers.
[1] https://github.com/MordorWide/EANation
AT1C Protocol AI Agents require Human consent. Don't trust us Verify the Receipt. https://www.npmjs.com/package/@at1c/sdk
oh you know, just ai-stuff
https://github.com/gabrilend/ai-stuff
https://slopshare.dev
Not really sure what to call it, Linux kernel ish implemented in JS so you can run it in the browser. Each program runs in WASM. Everything bootstrapped from a C/Cpp compiler compiled from source with a new target. Also kinda turned into Docker for WASM.
Recommended for people that feel a strong urge to run a browser inside a browser.
https://slopshare.dev/v0.1.5/wine/
Also quite useful to run coding agents sandboxed inside the browser, with HTTP policies and "backend managed skills/extensions". Compiles Codex from source (has a rust compiler), runs pi agent and CC (the old node one) on a custom javascript runtime on top of quickjs compiled inside the environment.
Everything is static JS (or the robots are lying to me)
Is it something that can run compiled WASI binaries? Like the kernel part that implements scheduling/fs/networking/gpu?
There's a commercial project like that called BrowserPod, but I also saw some open source implementations.
How does it compare to that?
It's very similar in concept to BrowserPod but with my own syscalls and some quirks I thought were cool. Not attempting to be Linux just attempting to be able to compile stuff that makes sense. Also it's wasm64 browserpod is wasm32.
For example I never added different users/permissions because it's simply not needed in this sort of env.
wasm64-unknown-dolly wasm32-browserpod-linux wasm32-unknown-emscripten
This means everything has to be recompiled from the ground up (starting from compiling the shell inside the environment, then the tools (ls, grep, etc.), then things like make and git, etc. etc.. Just compiled godot editor but doesn't run very well yet. Caching/packaging compiled stuff becomes a problem so some Docker ish syntax evolved from there. Ex: https://slopshare.dev/v0.1.5/view/zero-ad/
https://github.com/daugasauron/dolly
Also its just a project for myself to learn wasm stuff and push the limits of what it can do, but I feel more and more it's a very powerful concept useful in practice
I'm rebuilding a benchtop CNC milling machine.
Honestly? Still just trying to get hired.
But apart from that, I just updated my website's theme to something I wanted to do for quite a while now – a fantasy desktop! If you'd like to check it out: https://0xff.nu
I'm also trying to just learn new things with the time I was refunded by being unemployed.
Uh, well, I've been maniacally obsessed with building my own harness/factory/memory & context layer/build/ship/deploy pipeline because, well.. I like to automate things.
Take a look; https://useflower.dev
Since the 1.0 announcement for flet.dev I have been working on making small apps to make DXF files for CNC projects with drawn previews.
https://studio.flet.dev/run/bicQWc3Aul/
which is now at a working state for rabbet joinery boxes with a sliding lid, or omitting the lid:
https://community.carbide3d.com/uploads/default/original/3X/...
Next up is a hinged lid option and cleaning up the wireframe preview to look better/be more accurate.
working on apyhub mcp: to enable all apyhub apis to be available for agents. https://apyhub.com/mcp
I'm working on a new relational / logic language inspired by datalog and Bruce MacLennan's pioneering work on relational programming, but adopting the open-world assumption to allow missing values / heterogeneous data and enforce data integrity. I think logic languages will become increasingly important in the LLM era to provide strong correctness guarantees, and as a source of reinforcement learning. And I'm super fascinated by relational data models -- I think they have many advantages over the object-oriented view.
https://github.com/roland-redmount/atom https://roland-redmount.github.io/atom/
It's work in progress, many features are not completed. Any feedback is appreciated!
I'm working on two moderately large projects right now, both of which are interesting to me and maybe no one else, but why not share?
The first is withfire.app, which is basically a productization of the AI driven resume tailoring pipeline I used during my last job search. The landing page is still (AI) sloppy, but I think it has some neat features:
- a slick bookmarklet intake system for new job postings - just click the bookmarklet and the job is parsed automatically and added to your list
- improve-over-time tailoring: each tailoring will find skills the posting mentions that you don't have in your profile and ask you whether you have that skill, so your profile gets better and more specific over time
- nice, human-looking output options that look almost pixel for pixel like the hand-rolled resumes I used to submit
This system increased my hear-back rate for jobs by ~5x when I implemented it, and I figured it might be worth productizing. If you're interested in poking at please feel free to reach out! This is the first app I've ever built using AI end-to-end and it's been an awesome learning experience.
--
The second thing I'm working on is an astrophotography tool called ministar (https://github.com/galbacarys/ministar). The premise is basically, let's automate the hard parts of untracked astrophotography by using plate solving to help you point your camera to your target, and then live-stacking the result taking into account the rotation of the Earth. It's very early days and hacky as fuck, but i'm hoping to port it to Linux/Raspberry pi in the near future so that you can use just a DSLR, a 3 way head, a tripod, a power bank and an SBC for a minimal but high quality astronomy setup anywhere. I took my first shots of the h-Persei cluster last night and the results were really great, I'm hoping to get a longer shot at Andromeda later this week.
If you're interested in either of these and wanna chat, ping me at gabethebando (at) proton (dot) me, I'm kinda loving this brave new world where my side projects actually gain enough legs fast enough for me to stay interested in them
Trying to finish up a command line tool in go to interface with my Gramps Web service. Got really into genealogy and while ancestry, the Mormons, etc. Are helpful sources, I want to own my own data. LLMs are fantastic at genealogy work but there is a missing space for interaction with gramps via CLI. It's been vibed but cleaning up the code to make it worthy of being released publicly has proven a chore. We don't need more slop projects and I don't want to add to the problem.
https://pagey.site - A 100% free static site hosting for all your (and mine) vibe-coded slop. Truly 100% free. No strings attached.
As of now, it hosts more than 11K sites and 4.5K+ users within just 5 months of launch. It's been an honour to run it, truly.
I'm still running and improving it on almost daily basis.
Side hobby: Improving the gameplay of a multiplayer browser game. Rimpark (https://rimpark.gg) launched ~1.5y ago.
I'm self hosting a GLM 5.3 Flash, and with it: adding new mechanics, fixing bugs, etc.
It's been a learning journey so far: how to run inference, setup agentic systems, experimental loops, and prevent slop.
This week, I'm happy that my agent system can implement a new feature e2e with not a lot of human intervention in code reviews. Still reading every line, didn't YOLO (yet lol).
I "accidentally" created one of the fastest modern JavaScript frameworks out there: https://signy.dev/
...according to my own benchmarks (yeah, I know how that sounds).
Next, I'm planning to submit it to independent third party benchmarks, hoping to find out whether my results are accurate... or not :)
I've also just finished a CSS framework: https://iriscss.com/
Now I'm working on a new backend programming language that runs on the Erlang BEAM, along with a few web applications built using this tech stack.
I'm working on acoustic levitation using ultrasound.
Starting with a parabolic approach for this initial version, but the next version I'm working on uses a much more complex flat plane with phased array of 40khz ultrasound transmitters.
Just for research purposes, no real goal in mind other than learning about it.
I’ve been fooling around with a lot of consumer GPU hardware and having a fun time. There are so many rabbit holes and interesting parts available to piece a Rube Goldberg of a performant clanker together.
https://doug.sh/posts/six-gpus-one-x16-slot/
User documentation. Oof
d
Launched Diligly today. Investor-ready data rooms for startups.
An MIT re-implementation of an AGPL project with a very anti AI community.
getting rid of ai slop that has consumed the internet.
email, messaging, and social media with zero ai generated content.
www.channel.is
Ur is an (in-progress) universal formatter and linter for every programming language. It's in preview right now, which just means friends are messing around with it and having a good time pointing out all the bugs. Docs are mostly up to date, but buyer beware.
https://ur.dev/en/
I've been learning kubernetes with the help of Claude Opus and trying to build it into a course which others might find useful: https://gravitypersists.github.io/kubernetes-from-the-proble...
It's not meant to be super in-depth, more so aimed at understanding the high-level details and motivations of k8s. I figured that the best way to do that was to pose each chapter as an evolution of a startup experiencing a problem, and showing how k8s solves that problem, by treating the reader like the technical manager in the situation.
I'm not a k8s expert, so I can't definitively say it's "my course" but I will say, we're at an interesting time where you can contribute authorship to something and learn it at the same time. I'm spending more time steering the output the more I learn things, so it's not just slop.
A bloated club distro to make sure everyone has a standardized environment for automated testing frameworks, context specific documentation, and viable applications. =3
I'm so fucking tired. I built this tool, it's great, I love it, but its full capabilities are too complex to udnerstand. Because of that I struggle with marketing, as everyone is too lazy to try it out. Because of that I get no feedback, nothing to improve, and I can't make the tool any simpler, as otherwise it defeats the magic behind it.
Software is complicated by nature. It's our job to make it easy for the user.
Take a look at FFmpeg as an example. It's a complex tool that can do most things audio/video. A power user can do magical things, but they still provide simple commands for the most basic use-cases like conversion or trimming.
https://www.clipcat.com/blog/how-to-use-ffmpeg-a-beginners-g...
1. Provide sensible defaults 2. Study user journeys and write simple manuals for them. 3. Ask friends or internet communities for feedback.
Don't get discouraged, you can do it!
Pivoting into content creation. It's gonna be a series of videos of me programming, without AI assistance, so that some echo of the old ways persists, or at least is a little bit louder, for my having been here to preserve it. Mainly it will be work on personal projects I hold dear. I was inspired by the likes of Tsoding, although like Hayyoth I type into a box to narrate rather than speaking, as no one really wants to hear me blab.
Fixing a leaky skylight
Daily crossword puzzle app https://29words.com
I have been vibe coding with LLMs to see if the structure of latent space could be used to identify culturally bound concepts, like "fan death" and having it optimize runs of Qwen on my local GPU to test theories.
www.venndiagrammer.com
A traditional roguelike! Lots of fun and challenging myself to make sprites.
EV software. perf book. something edu/political related. a design for an upgrade to voting systems. ssh BBS game. mostly, EV software
interested in a performance book if you are sharing it somewhere
send an email to me and I'll let you know when the book, or early sample chapters, are avail
groglogic+perfbook@gmail.com
https://www.wherewhen.cc/, a travel planner, because nothing had everything I wanted. Wherewhen allows you to fix some times (eg a restaurant reservation) and moves everything else around that.
https://github.com/timbran-project/moor
Multiuser programmable persistent worlds. Compatible with LambdaMOO. Since 2022.
I'm working on decorating my homepage[1]. Other than that, I'm working on a language.
[1]https://www.makonea.com/en-US
Currently working on adding support for iPhone Duo to my hacker news app Hack. The app has been around for 9 years and when I made it, the menu bars and navigation buttons were custom views. That doesn't seem to work well for iPhone duo due to its vertical stack menu on the right side. So, I've been switching over to iOS native navigation bars and menu bars along with a bunch of other improvements. I was hoping to have it done couple weeks ago but I had to leave country for a family emergency. So I am working on it this week. It's mostly done, I'm just doing some final testing now and will submit to Apple this week:
https://apps.apple.com/app/id1464477788
Merge into one of the other copies:
(3 days ago /25p/56c) https://news.ycombinator.com/item?id=50389501
(7 days ago /23p/77c) https://news.ycombinator.com/item?id=50328340
Yes, it's on my list to automate these and I even reserved an account for this way back when: https://news.ycombinator.com/user?id=whoisworking.
Will merge the others.
Edit: ok, that's done and I did my best to normalize the timestamps for the posts from the older threads, but it's a bit messy. If you see weird things that's why.
I thought people were spamming this thread by making multiple posts about their project. Maybe pin a message that you've combined the threads?
Ah I had been intending to find any duplicates but ran out of time this afternoon. Let me see what I can do...
Edit: ok, I've attempted to remove the dupes now. In a few cases the same user posted comments about what appear to be different projects, and those ones I left.
infra engineering meanwhile trying to build some stuff for my own. also trying to finish cs master’s
Mostly trying to think about what career to pivot to, since SWE wages are going to get crushed in the next 5-10 years.
I've been working on online payment rails for businesses, developers and agents called Tuppence (https://tuppence.ai).
Accept online card payments from just 0.79% (UK) and earn 0.89% on all card spend [1], or get paid out to your bank.
I'm building tuppence to be available in the US, and 20 EU countries.
I'm getting a Point of Sale (POS) manufactured in the UK [2] for those who want to accept payments in person with card readers.
My goal is to help people sell their goods and make a living without paying over the odds for the privilege.
I charge a competitive fee for payment processing, give money back on card spend (with no cashback caps/limits) and do not charge saas fees or lock people into contracts.
Tuppences currently allows people to sell online with a store builder, expense management, payment links, direct debits & bookings - essentially Tuppence is a 'business' layer on top of what you get from the likes of Stripe, Adyen and Checkout.com. There's also have a 'connect' platform, so you can roll this out to your marketplace business.
Next on my roadmap is ticketing and 'treasury'. Treasury i'm personally excited about as you'll be able to earn 4% yield on your earnings, FDIC Insurance (FSCS in the UK), multi-currency banking in fiat balances such as USD, EUR, and GBP, or stablecoin balances like OUSD and USDC, and have bank account numbers to send or receive funds.
---
It's brand new, i'm still testing payment -> payout flows. These take a little time to verify on live payment rails.
Happy to give some fee-free payment processing for those who reach out and want to be early adopters, I can be reached at ollie@tuppence.ai, or comment on thread.
---
Tech Stack TLDR:
* TypeScript monorepo covering the API, dashboard, docs, mobile app and SDKs. * The API runs on Ts.ED over Express with TypeORM and Postgres. * Background jobs use pg-boss, so the queue lives in Postgres and there’s no Redis or separate broker. * The dashboard, till and storefront are React + Vite + Tailwind, the docs and help centre are Astro, and the mobile app is React Native/Expo. We ship SDKs for Node, Python, iOS, Android, React Native and WordPress, plus an MCP server and agent tools. * Everything runs on Google Cloud Run and Cloud SQL.
References:
[1] Card issuing: https://tuppence.ai/cards [2] Point of sale: https://tuppence.ai/pos/ridge/
How do you manage to keep rates down to 0.79%?
Another novel.
Currently, I'm in the midst of ricing my 3D printer - an Ender 3 V3 SE which has been sitting unloved for quite some time. Fresh filament, cleaning, mechanical tuneup, and now running through the very nice suite of calibration prints that are included in Orca slicer (which is new to me). It's been fun to get my hands around something physical.
Also been playing a bit of Blood Dungeon, which is pretty fun. I've also got a Factorio save that has mildly petered out after landing on Vulcanus (which I have kinda taken advantage of to break the addiction lol, but I'm sure I'll get back in there sometime over the winter).
I've been kinda deliberately taking a bit of a break from programming "side projects" after spending a humongous amount of free time and energy on a conference talk I gave a few weeks ago. It was titled "Human Creativity in the Age of AI" (which I later learned was an unfortunately generic title, lol), and I built this whole big musical and visual contraption that drove the presentation, including a "guitar solo generator" which recuts phrases of my own playing into a new solo on the fly.
It was a really rewarding project overall but definitely made me remember why I hadn't done public speaking for quite some time before - I am way too perfectionistic about it and it just becomes a massive time sink.
This one was worth it though, got a great reception and I'm really proud of the talk on the whole.
In the 2-3 weeks after the talk I spent a ton of time editing and laying out a really nice video of it, which I am also quite proud of. I didn't know anything about Davinci Resolve and I was able to get a really nice jump start with some help from Codex. I'm sure I don't know 98% of what a software package this big can do, but I did learn enough to be dangerous, and also enough to cut together some nice short clips of the talk as well.
It's been pretty rewarding to get a handle on a powerful bit of creativity kit like Resolve. It's incredibly capable and I'm really delighted with the results I was able to get. Obviously I am a rank amateur but to my untrained eye, my videos look pretty damn good. (And I could go on for another few paragraphs about how much work it took to get good results from the relatively shitty source video and audio I was working with).
I'm planning on doing some writeups because there was so much interesting tooling that came together as I worked on the talk and the videos, I could mine this one project for a year's worth of blog posts if I wanted to!
Anyway: links.
OrcaSlicer calibration guide - seriously great: https://www.orcaslicer.com/wiki/guides/calibration_guide
Blood Dungeon: https://store.steampowered.com/app/2431080/Blood_Dungeon/
I assume a link to Factorio is unnecessary here, ha.
My talk "landing page": https://epiccoleman.com/tripulator
Talk video: https://youtu.be/kD5qYiLIhJc
One of the shorts - the guitar solo part. I'm really stoked on the "presentation" bells and whistles in this video (though I think they're a bit mismatched to the regular intent of YouTube Shorts - analytics indicate that the intro part I slaved over is probably too long and people swipe away before the good stuff, but oh well).
https://www.youtube.com/shorts/5AE5uMv8VEs
quantum stuff
Guaranteed weight loss within 2–3 weeks; many people start seeing results within the very first week. 17,000+ members already!
Guaranteed results are achieved using the state of the art adaptive TDEE (total calorie burn of the day) tracking technique we've developed. It's sensorless, so no watch or other wearable is needed. It's way more accurate than those wearables (data included on the website), and our group has 2,000+ members ( https://macrocodex.app/community/ ). You can also read Play Store reviews.
All this completely free, no subscription or ads: https://macrocodex.app/
MacroCodex also works on both Android and iOS and any platform where PWA app can run.
Other other 2 apps are Android only (for now):
If you are looking for a workout app with Alpha Progression or MacroFactor Workout-like “Smart Progression,” or an app where you can run any program like Boostcamp but ad- and subscription-free, check out the Symbiote workout app (beta) https://symbiote-studio.macrocodex.app/
IF you are looking for AI photo Calorie tracking with lots of features which make it very fast to log food on daily basis checkout CalorieCodex (beta): https://caloriecodex.macrocodex.app/
All the apps have been under active development for a few months now.
.
Please don't delete and repost.
This is in both https://news.ycombinator.com/newsguidelines.html and https://news.ycombinator.com/newsfaq.html.
I'm working on a tool library for you, your apps and AI agents, it is on Github but super early though (alpha prerelease, web app and CLI etc coming very soon)
https://github.com/tool-io/tools-ts
The goal of https://tool.io is to be a Wikipedia but for functional software tools. Instead of recreating all kinds of small tools, use an open-source library instead, and you can contribute and create your own tools and publish them (by contributing on Github for now)
For years, I struggled to make any headway with Linux - I get so frustrated at how obscure, inscrutable, and unintuitive TUIs are! God I don’t want to waste my life typing esoteric strings of commands into the terminal. When I started with computers that was all there was, and once GUIs happened I never looked back.
Good news though, these days Claude will just do it for me!
I’ve been working on setting up a home media server, using headless Debian on a pi - I’ve got it set up to stream music and videos, hold all my ROMs, I’ve got it automatically standardizing file names and handling tagging and subtitles - all attached to a big 6-drive enclosure that I just stuffed all my random old HDDs into.
Even though I haven’t written more than a dozen lines of code or commands for the whole thing, I am pretty fuckin pleased that now I can just wave a magic wand and say “I want all my video game soundtracks filed under the game name rather than the artist” and Claude uh finds a way to grant my wish. I spend about an hour each night tweaking it and it’s so easy to customize to be exactly what I need.