i didnt use em atall for a long time but since recently i use them almost exclusively, they definitely passed my skills for larger projects...
i code by hand for fun, got few tinkering projects. to learn new things i can tell the thing. sometimes i get them from what the models output. play with it by hand to get a feel for it.
at all? i guess no one, but i think there's still a portion of people writing part of their code by hand, like libs, utils, classes etc that the agent can use directly instead the agent will generate their own piece from the scratch in order to protect maintainability
Yeah that’s interesting - I feel like smaller chunks of work that aren’t inherently complex like class definitions and utils etc are actually one of the nicest things LLMs help automate - I can understand a human modifying the output directly to finesse it to exactly what they want but I’d be curious why someone wouldn’t want to use an LLM to generate all the boilerplate and make a quick first draft of something that gets 80-90% of the way there for these kinds of things.
Definitely curious if anyone is actually doing this fully by hand and if so, why!
I do. I have tried using LLMs ( as a bootstrapped solopreneur tried optimizations via free AI ) it worked out ok for fast prototypes but even so had enough issues to not consider it for my non-throw away serious work (including iOS game in production) except in a limited fashion such as to spin up few shallow SwiftUI views , get typical implementation for stuff like IAP , GDPR regs, etc. for all the non-core code which is used by almost every game. This is what I used to look up Stack Overflow etc. pre AI. I do not turn on AI integration in Xcode. Leary of sharing my code base on my proprietary code for the game as well as my other products.
Zero usage of LLM (assembly coding, and sometimes plain and simple C).
I cannot access them to test if they would be of any help in my coding use cases.
You tell me once we get some inference access with a web API with public tokens (probably severely rate limited), or with full interop on noscript/basic HTML.
I may have to run locally open weight coding frontier models (slooooooow).
Very interesting - why aren’t you able to access them outside of a web API right now?
Have you tried smaller quantized open weight models that aren’t frontier? They probably can’t automate all your coding but I imagine they could at least help a lot with the drudge work that just takes a lot of time but isn’t necessarily complex?
Do you know of coding inference models I can access with curl? (namely with public access tokens, probably severely rate limited).
Or some coding inference models I can access with a noscript/basic HTML browser? (namely basic HTML forms)
If those inference models are still gated by whatwg cartel web engines, I will have to run full blown coding frontier models locally (if I want to have a chance at getting quality code). It is going to be very slow, and even slower while I am developping "prompt templates".
I did say "quality code", because I did ask some people already to generate classic and basic code paths using AIs, all were quite disappointing. That said, it was millions of years ago (in AI improvement time), namely a few months ago in human time :)
Commenting because I'm interested in the answers as well.
Stats say over 90% of engineers use AI to code, but the actual split of manual/AI is harder to say.
Sometimes I do on leetcode type things to keep my skills but today on the job I have run out of my AI allotment and Im not sure what to do.
Hahaha great point yeah that might be one of the few times people still need to code by hand
i didnt use em atall for a long time but since recently i use them almost exclusively, they definitely passed my skills for larger projects...
i code by hand for fun, got few tinkering projects. to learn new things i can tell the thing. sometimes i get them from what the models output. play with it by hand to get a feel for it.
at all? i guess no one, but i think there's still a portion of people writing part of their code by hand, like libs, utils, classes etc that the agent can use directly instead the agent will generate their own piece from the scratch in order to protect maintainability
Yeah that’s interesting - I feel like smaller chunks of work that aren’t inherently complex like class definitions and utils etc are actually one of the nicest things LLMs help automate - I can understand a human modifying the output directly to finesse it to exactly what they want but I’d be curious why someone wouldn’t want to use an LLM to generate all the boilerplate and make a quick first draft of something that gets 80-90% of the way there for these kinds of things.
Definitely curious if anyone is actually doing this fully by hand and if so, why!
I do. I have tried using LLMs ( as a bootstrapped solopreneur tried optimizations via free AI ) it worked out ok for fast prototypes but even so had enough issues to not consider it for my non-throw away serious work (including iOS game in production) except in a limited fashion such as to spin up few shallow SwiftUI views , get typical implementation for stuff like IAP , GDPR regs, etc. for all the non-core code which is used by almost every game. This is what I used to look up Stack Overflow etc. pre AI. I do not turn on AI integration in Xcode. Leary of sharing my code base on my proprietary code for the game as well as my other products.
I do, not quite every single day.
Amazing. What do you still code by hand and what do you leave to the LLMs now?
how many calculate numbers on pen and papers. Still
I’m actually curious what the answer is to this too lol
Zero usage of LLM (assembly coding, and sometimes plain and simple C).
I cannot access them to test if they would be of any help in my coding use cases.
You tell me once we get some inference access with a web API with public tokens (probably severely rate limited), or with full interop on noscript/basic HTML.
I may have to run locally open weight coding frontier models (slooooooow).
Very interesting - why aren’t you able to access them outside of a web API right now?
Have you tried smaller quantized open weight models that aren’t frontier? They probably can’t automate all your coding but I imagine they could at least help a lot with the drudge work that just takes a lot of time but isn’t necessarily complex?
Do you know of coding inference models I can access with curl? (namely with public access tokens, probably severely rate limited).
Or some coding inference models I can access with a noscript/basic HTML browser? (namely basic HTML forms)
If those inference models are still gated by whatwg cartel web engines, I will have to run full blown coding frontier models locally (if I want to have a chance at getting quality code). It is going to be very slow, and even slower while I am developping "prompt templates".
I did say "quality code", because I did ask some people already to generate classic and basic code paths using AIs, all were quite disappointing. That said, it was millions of years ago (in AI improvement time), namely a few months ago in human time :)
[dead]