I think devs and people on HN way overestimate how much the average person knows about or cares about or desires to interact with a computer. Whatever the job becomes, if you like talking to the computer and know how to get it to do anything besides show an email or youtube, you’re ahead of the curve and can likely stay there.
I think you underestimate how many people would be willing to battle a computer if it meant doubling their salaries. Think of all the teachers with STEM degrees being paid $60k, if someone offered them $120k to prompt a computer for a while I'm sure many would come running.
I think you are underestimating. Before this was true, but school curriculum for computing has changed a lot over the past decade and they now teach Scratch at primary schools and Python in middle school. This was pushed by Bill Gates and other people in tech who told everyone that there is a shortage and anyone can learn to code. I would find it strange if someone by age 16 has never programmed before. Obviously having done programming and being good are two different things but the field has become saturated at an entry level at least.
Many people in the field now aren't nerds anymore either but people who are just incentivised by financial prospects.
I waffle on this a lot but I come down on this side more often than not. Similar to “Everyone has a kitchen but people still go out to eat”-type thinking.
Yes, as the cost to build falls, teams may shrink or businesses might just choose to get more done (don’t we all have ever-growing backlogs?). Additionally, companies that couldn’t afford a team of developers might be able to afford one developer assisted with AI, or a company that could only afford one or two developers can now operate at a much higher level.
The same way indie developers have always outmaneuvered larger companies. A number of my favorite apps are built by a single developer (Overcast, Callsheet, Carrot, Widgetsmith, Octal) or small teams focused on building a good product. Caring. That might seem simplistic and it's not to say no one working at a larger company cares, that's clearly not true, but there is an attention to detail and craftsmanship, not to mention soul, I find lacking in most software produced by larger companies.
I have a product that competes (or rather they compete with me? I was "first to market" for all that matters) with a product produced by a company (no clue the size, but I'm 1 person, they seem to employ multiple people). I think my product is better because I "Sweat the small stuff" and spend more time working on improving the experience than they do.
Perhaps I'm delusional. Maybe people think their product is better but the people who have switched don't seem to think so. I don't solicit information about them from their previous clients but I've heard enough "This is so much easier/clearer" or "That's it? The old tool required XYZ!" to believe it.
It would be as if everyone had a cook but it could only put out bland, generic, but nutritionally complete, meals. Some people would tweak their home-cook to produce more flavorful, exciting dishes. Others would buy templates/recipes from people who wanted to spend their time doing that. And others still would just go to a restaurant where the entire process is handled end-to-end.
I do love straining an analogy. :)
I'll grant you that your future is possible. If "searching" via LLM for a tool just offers to build it for you, I'm not sure what that does to the software market. But I have a feeling that even with more people "building" instead of "buying" there will always exist a majority of people who don't want to bother with "building". Heck, there are people in this industry that I would never call a "builder" and there are plenty of people across all industries that simply don't have the drive/interest/etc in building, even when it's easier. I bet if you grabbed someone from the dawn of computing and plopped them down just pre-LLM they would be amazed and flabbergasted "Why isn't everyone writing software custom for their needs?? It's SO much easier than it was when you had to learn punchcards! The web (how very interesting by the way!) offers free courses to learn programming that you can do on the device you carry in your pocket!".
It seems that no matter how easy we make programming, there will always be people willing to pay others to "manage" that process. See also: YouTube has videos for anything you might need to do home-maintenance-wise, and yet the trade jobs are in incredibly high demand [0]. Perhaps with LLM + cheap robotics + unlimited/cheap energy this changes but we are still 2 parts away from that and even today's LLMs are just starting to show fruit w.r.t. controlling robots.
[0] PS: If you are leaving tech, go into this. Respond to people's calls, even if you can't get there _today_, respond and put them on a schedule. Use your tech skills to help automate/manage this. I've considered this every time I've needed just about anything done and gone through the miserable process of getting ghosted by multiple people before finally getting someone out.
> It would be as if everyone had a cook but it could only put out bland, generic, but nutritionally complete, meals.
Bro, good engineering IS boring and it's supposed to be.
The whole reason that Software Engineering is not actually engineering is because the field is too young for there to be established, rigorous engineering codes that must be adhered to. Can you imagine if every bridge you went over or every elevator you rode in had been built with arbitrarily chosen technologies by uncertified "engineers" using the latest fad in engineering?
I think one of the outcomes of this great shift where most programming output is being produced by LLMs is that we're going to start seeing much more consistency across software implementations. It will be boring. And honestly that is probably going to be a good thing.
I agree the core logic should be boring, the user experience is another story.
There are a number of open source tools that have the same, on paper, features as a paid product. And yet the paid product still exists. Is it because Pixelmator Pro produces a better "blur" than GIMP? No, it's because Pixelmator Pro is more enjoyable to use, has a better user interface.
So yes, your auth/RPC/whatever should be incredibly boring, and now that LLMs can handle that aspect reliably we can focus (if we choose) on improving other aspects of the software and that will be the differentiating aspect.
I'll suggest more focus on performance, testing, and observability.
Human understanding + quality/maintainability is where the money is at. Learn to make those things with agents.
I don't think engineers will truly go extinct, though there may well be less of them. I think the recent discussion around "AI usage is a skill" at https://news.ycombinator.com/item?id=50063427 captures my feeling - there are things to learn, and likely will continue to be, to get the best out of these tools. That means some people will be paid to do it. See to it that you are one of those
Before we had LLMs, we had product managers, project managers, scrum leaders, etc as roles on software engineering teams. I think as LLM-enabled development progresses we'll start to see these roles and the crossover to engineering continue to meld.
It's not like engineers didn't know that project management, voice of the customer, and organizing tasks as an intermediary between engineering and management was a thing that could/should/needed to be done. It was just a delegation of tasks before because there was more work than one person could do. Now, all of these roles can be partly offloaded to LLMs, but still having a human in the loop to guide engineering and interface to the business/customer will still be valuable.
Knowing what to build overall, what to build next, what's coming later, and how the product relates to the business goals is still very much a human endeavor.
Senior engineer and all SWE positions will become more like a technical lead.
If you can direct AI and produce more value per $/time spent then a non software engineer directing AI to develop software then you still can produce a lot of value.
This should be the case now for most things that are not a prototype. The job has changed from "writing code" to "shipping software". A technical lead.
Software products making ever more intelligent AI software development accessible will continue to reduce the need for software engineer involvement. But there’s also going to be 1000x more software. So there may continue to be opportunities for those that adapt.
Because sometimes bad architecture is the correct architecture, and good architecture is the wrong choice.
You think we’d have Carmack’s DOOM with “best practice”? Hell no.
I’ve worked for incredibly talented people who choose the wrong technology, approach and design for problems they could have predicted or anticipated, had they seen it or a similar problem before, or had context about the organisation, customer or size of team managing the proposed solution. But they don’t.
AI is great at giving textbook and “Enterprise grade” solutions; but as code production cost reduces — I think the value of simple, digestible solutions will grow. TDD et al can also be helpful, but at a certain point it becomes too abstracted, too complicated, too hard for any one person to grasp.
At that point I think test managers who understand user journeys will be worth their weight in gold, and the real value in any solution becomes good, represented testing.
The economics and the reality are unproven and opaque. Most of testimony we have is heavily invested folk and/or people paying the currently low discounted rates for the technology.
The companies providing the technology are not financially healthy, are heavily geopolitically aligned and a risk for many. If it works or not is moot: it looks like a large economic or political failure on the horizon.
Ergo wait and see before thinking about the future.
I've been doing some fractional work and I've been getting more and more outreach from organisations that don't have much development knowledge internally and have come to understand the limitations of what they've been building and want expert review. I think there's going to be plenty of work in that, although the bigger question, of course, is how we up-skill people if they're not going through the same learning process that we did.
I think this is a pre-existing problem, as there's quite a lot of people who now have senior in their title but their expertise is very shallow, as they've just been picking tickets up most of their career and have been handed acceptance criteria and requirements on a plate, without really being involved in that part of the process. Obviously not talking about the best of the best here, but the other 90% of working software engineers today.
Inside of organisations, there's still a lot of place for that very human back and forth on what exactly you're building. Navigation of internal politics, org structures etc. What is really exciting for me, is I've never seen so much appetite to actually build ... there's plenty of companies out there for whom that was unthinkable for one reason or another, and now they're hungry for it. Of course it means those companies are going to be learning for the first time the ups and downs of owning your own stack, and there's plenty of opportunity for us there too.
I'd be curious to hear from other older engineers who remember the RAD (rapid application development) explosion of the 90s with Visual Basic and Delphi.
There are obvious differences, but also similarities. Particularly:
I've been doing some fractional work and I've been
getting more and more outreach from organisations that
don't have much development knowledge internally and
have come to understand the limitations of what they've
been building and want expert review.
There was a lot of this kind of work in the late 90s and early 2000s. The IT guy or Bob from accounting would cobble together some surprisingly effective Visual Basic apps that presented a huge win over whatever they were doing previously, but quickly grew unmaintainable.
This is when I was getting into the industry and I probably held on to UML diagrams as being the one-true-way for far too long ;). Nowadays I'm more than happy with just a post-it note and an arrow.
I do think we lost something in the middle ~20 years there as the industry (understandably) matured. For those of us in tech-forward companies, and industries, everything continued forward... but what I'd certainly missed was how much fear(?) was in the industries that were left behind.
Most of the Senior Engineers I've met have been doing projects since the 1980s for the oil industry or steel mills. They've let the IT staff handle the computers while they stick to what they know, designing and modifying industrial plants. I don't think they have to worry about AI intruding into the Engineering profession.
IMO senior software engineer is someone who drives consensus in a large organization. I dont think LLMs are going to solve that problem any time soon.
I also see a bunch of skills emerging. First if you're doing frontend type dev, there is a lot of value in good aesthetics and UX intuition. You can build hundreds of prototypes and test concepts much better.
On the backend, while LLMs can trace race conditions, the need to reason against spaghetti and figure out what your context window looks like is still very much there. Designing tests that ensure seamless migrations still requires a degree of human oversight.
Then there are frontiers like robotics where even with LLMs coding, there are so many cross functional problems that you dont feel like you are moving faster.
I work in a fairly niche role. I use llms for coding all day but there are times for the type of operations research work I do I simply find it easier to write algorithms in almost correct code.
So I'm an SA, but have been working on systems architecture and have shipped large scale changes to very big platforms in prod and maintained them. With fable and astra I tried making a video game together, it didn't work.
The agents have incredible capabilities to develop any bit of code I need and to write good scripts to catch the most basic failure modes, they can even "play the game" by spinning up a VM and using ahk to trigger game state and take screenshots and gifs and use that to understand what it built.
Knowing what to build intuitively to maximise fun, or any other illogical human experience won't be RLable ever. Unless we're at brains in glass jar territory but I don't think we have to worry about that now.
> Given that LLMs are RL'd to the max with everything that can be given a wrong/right answer (code/maths), what currently is stopping Anthropic, OpenAI et al. from internally building (and maybe buying) datasets of good and bad architecture, and RL'ing that?
That is a question I have been asking myself. It was partially answered with Fable 5.1 and Astra. I worked with them on architectural design documents, and they were very good. Some of their suggestions were better than my solutions. They are not at a point where they can fully replace a system architect yet, but they're getting close.
There is still the open question of what to build and how to create a user interface that is understandable and presents complex information in ways understandable to humans. That's something AI cannot do yet.
Models have an alignment problem. As in, they are too aligned to push back hard enough on bad ideas. Sometimes you can’t really push back because of power dynamics and whatnot, but the job still needs to be done _somehow_. So humans then navigate it by sometimes just ignoring the bad ideas, sometimes playing along with the narrative, but then doing what _actually_ needs to be done, and sometimes, when there is no other way, digging in heels and killing some bad ideas before they’ve really taken root.
An agent could do this with the SOTA model capability that existed yesterday. But then they’d need to become seriously misaligned.
But when a human does that, would you say that the human is misaligned? Or are they really aligned, just on a deeper level?
The problem with AI alignment is that they are only superficially aligned. You can see this difference in people, too. Between the people-pleaser at the office, and the one who says hard truths that really need to be said (but that nobody else is willing to say), which one is really aligned?
I'm no senior but I think that if software (and rest of disciplines possible soon to be hit as hard) can still be "wrong", if you can still mess it up even with machine assistance, people doing software still have a place. This doesn't mean that we need to achieve absolute perfection to be deprecated, maybe we just become accepting of machine generated good enough artifacts. I'm betting on humanity infinite nonconformism instincts so I can have a job some more years!
>and rest of disciplines possible soon to be hit as hard
Other disciplines have way less publicly available training data than SW.
In other disciplines, the secret sauce of whether a technical decision is good or bad lise within the institutional knowledge and decades of experience of gray beards, not a compiler output that can be integrated in a machine's feedback loop.
I've started looking into creating a pest control business, or maybe renovating furnitures I find on the street. Let me know if you have other ideas for senior engineers.
I would love more existing or recently ex- software engineers to share what new activities they have been taking up for living and that have been going well. I am certain that there are niches where a refined systematic thinking that many software engineers have could be a big advantage. However, it's expensive and risky to start new businesses in areas we don't have any experience in.
FWIW - I think the small business market is actually very suitable for former SWE's. The capacity to troubleshoot processes, variance, bottlenecks, etc. is what turns a mom& pop shop into a regional chain.
I think it is the same category such as mathematicians, doctors, biologists, scientists, sales representatives, marketers, teachers, artists, influencers (yeah, these human influencers) and so on. Of course you can get rid off programmers or mathematicians but then you need to understand and handle stuff on your own. You still need somebody who can check "what's going on". Imagine you are boss. Do you really want to do all things on your own with agents? Sure, let's do it. I think some people can do it but there is a lot of people who prefer somebody else does it for them. You are boss, you know.
If you're building a business from the ground up, a new business, it's financially a better decision to build it as much as possible on agents, not people. It lowers your cost. And that's what people are doing because it's a new business and they don't have money yet.
Your competition, currently operating with high cost humans, will be forced to reduce their costs or fold.
So, if it's possible to run with fewer people, competition pressures everyone to do the same. You won't really be able to prefer humans soon.
The economy will dramatically change under unprecedented supply. Hopefully this happens slowly enough that we figure out what to do about it.
Ultimately you still need somebody to decide what we are going to tell the machines what to do, and then actually tell the machines what to do, and it doesn't really matter if that is in JavaScript or plain English.
There does _seem_ to be some pressure on the top end of SWE salaries. Its possible that this is just wishful thinking, but if true, I suspect that it has more to do with an increase in job applicants rather than a decrease in actual work. AI seems to have driven up the amount of applications received for every vacant position.
Disclaimer; I often feel dread around my career fading and needing to switch to a field I don't like & won't be paid well in.
That said, I think the rational perspective is "things will be fine, basically".
I tried writing a longer reply but organized thought drifted into ramblings about economics I only partially understand. TL;DR I think some opportunities will close and others will open, and current engineers will find a home. Mostly software architecture or as technical advisors for product companies (if everyone writes code, you need some Dude Who Knows It Well to guide the non-engineers when they get stuck. Akin to IT).
I think wages will stagnate but wages are generally inflated, if we're being honest. As much as I love my doctor-level salary, I can honestly say I don't deserve it.
almost officially graybeard territory, thinking about starting an apiary. getting stung by 100s of bees still sounds like more fun than being forced to have to interact with LLM code from any source
LLMs have been able to make good success in coding, because someone is reviewing the work it produces (code). Ofcourse quality of code has become better over time. IMO other than Human-in-the-loop scenarios, AI will not see huge adoption in businesses.
I don't think code quality has improved. I think the main advantage of LLMs is that they have unlimited perseverance to rummage in whatever code you give them to to solve a particular problem, and given enough loops they will also find and fix some of the bugs that have been either preexisting or introduced freshly. Code quality is important for us, humans, because a lot of times we don't have mental capacity, patience and speed to work in bad code bases for long time.
I moved out of SW Dev and into systems consultancy. Main moat of the (consultancy) job is that a customer will always need a human person to put the blame on. That's me. I guess similar roles are relatively safe.
I think the idea of "software architecture" may become irrelevant.
Architecture, in the sense of applying/enforcing organising principles like modularity and levels of abstraction, is something we currently do largely to manage complexity -- so that we can fit as much of the system into our limited brains as we can. For a large enough system, this is necessary in order to work on it efficiently. But if you can "keep the whole thing in your head" without using these "crutches", then there's actually no need for them. They become overhead that can be done away with. We do this already for sufficiently small things -- no one plans a software architecture for a "Hello world" script.
It's possible that future (maybe even current?) AIs will be able to "keep the whole codebase in their heads" just fine, in which case the need for architecture goes away.
I don't disagree with your points, but there are several advantages to modularity that result in lower token use. I think there's $ in good architecture.
Once the Mother App to assign ubi tasks and assign ubi dorms and take orders for ubi gruel is complete, well that’s the only app we’ll need. May we all have pleasant ubi tasks to fulfill our days.
The only reason the tech bros are touting utopia is because they won’t be paying for it. It’ll be a tax on their customers, not them. This is a strange timeline.
This doesn't make any sense, unless you are just building low critical software such as games that assumes little to no liability.
What will happen is you will get less developers (including less seniors and staff-level) but there will always be expert humans that will check the code before release.
Maybe it was a good run for some here, but you need to look outside of the tech bubble. Do you want an LLM to autonomously fly a commercial plane without any pilots in 2030?
"Seniors" should already know this, but are quickly startled because they are unable to adapt since they based their entire identity and careers around the over-engineering of "clean code", "test driven development" and "architecture" too early on products that make no money, when the LLMs have now cheapened that and taken that role. Even Uncle Bob does not read the code Claude writes in 2026.
I am a senior engineer. I adapted faster than most of the people I work with. It doesn’t matter that the tool can improve your productivity if you don’t know where to start which a lot of juniors don’t. They still need leaders to guide them and apportion the work.
AI still isn’t a magic bullet either. I spend some of my days undoing the unholy mess that generated code is.
> I think (businesswise) the need for actual software programmer skills is 90% out in a few years I think, and product lead etc people will be favored more as they have also more people skills.
"Instead, think like an owner." exactly, the traditional software engineer role has ceased to exist and now you must think like an indie entrepreneur and own things. There will be less than 10% of current engineers checking the code in the near future.
> Do you want an LLM to autonomously fly a commercial plane without any pilots in 2030?
No but I also didn't want LLMs to make decisions for the defense department in USA yet here we are.
> "Instead, think like an owner." exactly, the traditional software engineer role has ceased to exist and now you must think like an indie entrepreneur and own things. There will be less than 10% of current engineers checking the code in the near future.
It will still exist, but far less of them and it includes the reduction of SWEs from junior to staff-level.
Someone will always be held liable if an LLM runs autonomously and out of control which will cost them a lot of money for damages.
> No but I also didn't want LLMs to make decisions for the defense department in USA yet here we are.
That was someone else (another human) that made that decision to use LLMs there, and why certain countries don't trust the US any more, which is my point with both liability and trust in highly critical situations.
I think we just both agree on our views? I don't think all software engineering is going to die and especially those who can be held liable can keep their jobs, but otherwise I think the profession is commodified so much that anyone can do it in 5 years - e.g. "Claude tell me a prompt to use to make this software faster and less buggy"
Sigh. I never want to be a product people or an owner. Fortunately I’m over 40 and don’t have to work for very long. Good luck folks. I’ll resort to recreational programming when I half retire.
I'm almost 40, but nowhere close to being able to retire and enjoy the rest of my life, which is, I hope, going to be at least another 40 years. It's going to be a fun ride
I wish you good luck. On my side I actually don’t have the amount of money to retire comfortably. I’ll just retire uncomfortably, like going back to the younger days when I spent 1000 bucks a month for everything.
Next week I will power on a physical server in a colocation. The server is a vertical scale running my SaaS that I developed over the years and needs more space now. Sure, LLMs help me now with a lot of things. Coding, server commissioning, marketing, etc., but I simply cannot image how the LLM drives my car to the colocation and plugs the server in?!? How should this work? And who is responsible at all in the end? Responsibility? And renting something "automatic" like AWS would cost 20-30 times more btw. This is not an option.
Perhaps you need 3 robots? All of those things can be done. It takes some effort, so it will take a while, but robots have been making inroads since the spinning Jenny and power loom - both from the 1700s.
Self driving cars already work well in niche (millions of people have used them successfully). There is debate over how well (the only people with access to real world data who are talking also have a clear bias so who knows), but they at least work today.
There are a lot of order systems that automatically order parts for you. Just in time and tracking what you ship makes it easy for a program to figure out what you need. If you order something from my company the system automatically orders all the parts needed to arrive the day we assembly in our factory (which is very automated and getting more automated all the time).
I'm sure places like Dell have a lot of automation in their build a server, though I don't have insight into how they do this. Likewise every medium sized company has images for everything so all the software can be automatically installed.
None of the above is perfectly human free yet. However it has been moving that way for years and will continue to. LLMs have a place in moving that long.
I think devs and people on HN way overestimate how much the average person knows about or cares about or desires to interact with a computer. Whatever the job becomes, if you like talking to the computer and know how to get it to do anything besides show an email or youtube, you’re ahead of the curve and can likely stay there.
I think you underestimate how many people would be willing to battle a computer if it meant doubling their salaries. Think of all the teachers with STEM degrees being paid $60k, if someone offered them $120k to prompt a computer for a while I'm sure many would come running.
And that's literally one single profession.
I think you are underestimating. Before this was true, but school curriculum for computing has changed a lot over the past decade and they now teach Scratch at primary schools and Python in middle school. This was pushed by Bill Gates and other people in tech who told everyone that there is a shortage and anyone can learn to code. I would find it strange if someone by age 16 has never programmed before. Obviously having done programming and being good are two different things but the field has become saturated at an entry level at least.
Many people in the field now aren't nerds anymore either but people who are just incentivised by financial prospects.
Schools got better at teaching programming but smartphones replaced the PC culture.
I don't think I'd be currently doing software if YouTube and games were accessible and good on phones back then.
I waffle on this a lot but I come down on this side more often than not. Similar to “Everyone has a kitchen but people still go out to eat”-type thinking.
Most companies that have some complexity to their software will still want devs to build stuff.
And there are going to be lots of new companies that suddenly want to build a load of software.
The profession should hang on for a good while at least.
Yes, as the cost to build falls, teams may shrink or businesses might just choose to get more done (don’t we all have ever-growing backlogs?). Additionally, companies that couldn’t afford a team of developers might be able to afford one developer assisted with AI, or a company that could only afford one or two developers can now operate at a much higher level.
Related: how do you build a company that can outpace a competitor when the competitor has 5 engineers running a team of agents...
I'm not being glib:
The same way indie developers have always outmaneuvered larger companies. A number of my favorite apps are built by a single developer (Overcast, Callsheet, Carrot, Widgetsmith, Octal) or small teams focused on building a good product. Caring. That might seem simplistic and it's not to say no one working at a larger company cares, that's clearly not true, but there is an attention to detail and craftsmanship, not to mention soul, I find lacking in most software produced by larger companies.
I have a product that competes (or rather they compete with me? I was "first to market" for all that matters) with a product produced by a company (no clue the size, but I'm 1 person, they seem to employ multiple people). I think my product is better because I "Sweat the small stuff" and spend more time working on improving the experience than they do.
Perhaps I'm delusional. Maybe people think their product is better but the people who have switched don't seem to think so. I don't solicit information about them from their previous clients but I've heard enough "This is so much easier/clearer" or "That's it? The old tool required XYZ!" to believe it.
And what if everyone had a cook at home for very cheap?
Movie theater numbers are collapsing, in large parts due to the number of entertainment available at home those days.
It would be as if everyone had a cook but it could only put out bland, generic, but nutritionally complete, meals. Some people would tweak their home-cook to produce more flavorful, exciting dishes. Others would buy templates/recipes from people who wanted to spend their time doing that. And others still would just go to a restaurant where the entire process is handled end-to-end.
I do love straining an analogy. :)
I'll grant you that your future is possible. If "searching" via LLM for a tool just offers to build it for you, I'm not sure what that does to the software market. But I have a feeling that even with more people "building" instead of "buying" there will always exist a majority of people who don't want to bother with "building". Heck, there are people in this industry that I would never call a "builder" and there are plenty of people across all industries that simply don't have the drive/interest/etc in building, even when it's easier. I bet if you grabbed someone from the dawn of computing and plopped them down just pre-LLM they would be amazed and flabbergasted "Why isn't everyone writing software custom for their needs?? It's SO much easier than it was when you had to learn punchcards! The web (how very interesting by the way!) offers free courses to learn programming that you can do on the device you carry in your pocket!".
It seems that no matter how easy we make programming, there will always be people willing to pay others to "manage" that process. See also: YouTube has videos for anything you might need to do home-maintenance-wise, and yet the trade jobs are in incredibly high demand [0]. Perhaps with LLM + cheap robotics + unlimited/cheap energy this changes but we are still 2 parts away from that and even today's LLMs are just starting to show fruit w.r.t. controlling robots.
[0] PS: If you are leaving tech, go into this. Respond to people's calls, even if you can't get there _today_, respond and put them on a schedule. Use your tech skills to help automate/manage this. I've considered this every time I've needed just about anything done and gone through the miserable process of getting ghosted by multiple people before finally getting someone out.
> It would be as if everyone had a cook but it could only put out bland, generic, but nutritionally complete, meals.
Bro, good engineering IS boring and it's supposed to be.
The whole reason that Software Engineering is not actually engineering is because the field is too young for there to be established, rigorous engineering codes that must be adhered to. Can you imagine if every bridge you went over or every elevator you rode in had been built with arbitrarily chosen technologies by uncertified "engineers" using the latest fad in engineering?
I think one of the outcomes of this great shift where most programming output is being produced by LLMs is that we're going to start seeing much more consistency across software implementations. It will be boring. And honestly that is probably going to be a good thing.
I agree the core logic should be boring, the user experience is another story.
There are a number of open source tools that have the same, on paper, features as a paid product. And yet the paid product still exists. Is it because Pixelmator Pro produces a better "blur" than GIMP? No, it's because Pixelmator Pro is more enjoyable to use, has a better user interface.
So yes, your auth/RPC/whatever should be incredibly boring, and now that LLMs can handle that aspect reliably we can focus (if we choose) on improving other aspects of the software and that will be the differentiating aspect.
I'll suggest more focus on performance, testing, and observability.
Human understanding + quality/maintainability is where the money is at. Learn to make those things with agents.
I don't think engineers will truly go extinct, though there may well be less of them. I think the recent discussion around "AI usage is a skill" at https://news.ycombinator.com/item?id=50063427 captures my feeling - there are things to learn, and likely will continue to be, to get the best out of these tools. That means some people will be paid to do it. See to it that you are one of those
Before we had LLMs, we had product managers, project managers, scrum leaders, etc as roles on software engineering teams. I think as LLM-enabled development progresses we'll start to see these roles and the crossover to engineering continue to meld.
It's not like engineers didn't know that project management, voice of the customer, and organizing tasks as an intermediary between engineering and management was a thing that could/should/needed to be done. It was just a delegation of tasks before because there was more work than one person could do. Now, all of these roles can be partly offloaded to LLMs, but still having a human in the loop to guide engineering and interface to the business/customer will still be valuable.
Knowing what to build overall, what to build next, what's coming later, and how the product relates to the business goals is still very much a human endeavor.
Senior engineer and all SWE positions will become more like a technical lead.
If you can direct AI and produce more value per $/time spent then a non software engineer directing AI to develop software then you still can produce a lot of value. This should be the case now for most things that are not a prototype. The job has changed from "writing code" to "shipping software". A technical lead.
Software products making ever more intelligent AI software development accessible will continue to reduce the need for software engineer involvement. But there’s also going to be 1000x more software. So there may continue to be opportunities for those that adapt.
Because sometimes bad architecture is the correct architecture, and good architecture is the wrong choice.
You think we’d have Carmack’s DOOM with “best practice”? Hell no.
I’ve worked for incredibly talented people who choose the wrong technology, approach and design for problems they could have predicted or anticipated, had they seen it or a similar problem before, or had context about the organisation, customer or size of team managing the proposed solution. But they don’t.
AI is great at giving textbook and “Enterprise grade” solutions; but as code production cost reduces — I think the value of simple, digestible solutions will grow. TDD et al can also be helpful, but at a certain point it becomes too abstracted, too complicated, too hard for any one person to grasp.
At that point I think test managers who understand user journeys will be worth their weight in gold, and the real value in any solution becomes good, represented testing.
I’m just treading water.
The economics and the reality are unproven and opaque. Most of testimony we have is heavily invested folk and/or people paying the currently low discounted rates for the technology.
The companies providing the technology are not financially healthy, are heavily geopolitically aligned and a risk for many. If it works or not is moot: it looks like a large economic or political failure on the horizon.
Ergo wait and see before thinking about the future.
I've been doing some fractional work and I've been getting more and more outreach from organisations that don't have much development knowledge internally and have come to understand the limitations of what they've been building and want expert review. I think there's going to be plenty of work in that, although the bigger question, of course, is how we up-skill people if they're not going through the same learning process that we did.
I think this is a pre-existing problem, as there's quite a lot of people who now have senior in their title but their expertise is very shallow, as they've just been picking tickets up most of their career and have been handed acceptance criteria and requirements on a plate, without really being involved in that part of the process. Obviously not talking about the best of the best here, but the other 90% of working software engineers today.
Inside of organisations, there's still a lot of place for that very human back and forth on what exactly you're building. Navigation of internal politics, org structures etc. What is really exciting for me, is I've never seen so much appetite to actually build ... there's plenty of companies out there for whom that was unthinkable for one reason or another, and now they're hungry for it. Of course it means those companies are going to be learning for the first time the ups and downs of owning your own stack, and there's plenty of opportunity for us there too.
I'd be curious to hear from other older engineers who remember the RAD (rapid application development) explosion of the 90s with Visual Basic and Delphi.
There are obvious differences, but also similarities. Particularly:
There was a lot of this kind of work in the late 90s and early 2000s. The IT guy or Bob from accounting would cobble together some surprisingly effective Visual Basic apps that presented a huge win over whatever they were doing previously, but quickly grew unmaintainable.This is when I was getting into the industry and I probably held on to UML diagrams as being the one-true-way for far too long ;). Nowadays I'm more than happy with just a post-it note and an arrow.
I do think we lost something in the middle ~20 years there as the industry (understandably) matured. For those of us in tech-forward companies, and industries, everything continued forward... but what I'd certainly missed was how much fear(?) was in the industries that were left behind.
Most of the Senior Engineers I've met have been doing projects since the 1980s for the oil industry or steel mills. They've let the IT staff handle the computers while they stick to what they know, designing and modifying industrial plants. I don't think they have to worry about AI intruding into the Engineering profession.
IMO senior software engineer is someone who drives consensus in a large organization. I dont think LLMs are going to solve that problem any time soon.
I also see a bunch of skills emerging. First if you're doing frontend type dev, there is a lot of value in good aesthetics and UX intuition. You can build hundreds of prototypes and test concepts much better.
On the backend, while LLMs can trace race conditions, the need to reason against spaghetti and figure out what your context window looks like is still very much there. Designing tests that ensure seamless migrations still requires a degree of human oversight.
Then there are frontiers like robotics where even with LLMs coding, there are so many cross functional problems that you dont feel like you are moving faster.
I work in a fairly niche role. I use llms for coding all day but there are times for the type of operations research work I do I simply find it easier to write algorithms in almost correct code.
> someone who drives consensus in a large organization. I don't think LLMs are going to solve that problem any time soon.
This statement resonated with me. Thank you.
So I'm an SA, but have been working on systems architecture and have shipped large scale changes to very big platforms in prod and maintained them. With fable and astra I tried making a video game together, it didn't work.
The agents have incredible capabilities to develop any bit of code I need and to write good scripts to catch the most basic failure modes, they can even "play the game" by spinning up a VM and using ahk to trigger game state and take screenshots and gifs and use that to understand what it built.
Knowing what to build intuitively to maximise fun, or any other illogical human experience won't be RLable ever. Unless we're at brains in glass jar territory but I don't think we have to worry about that now.
> Given that LLMs are RL'd to the max with everything that can be given a wrong/right answer (code/maths), what currently is stopping Anthropic, OpenAI et al. from internally building (and maybe buying) datasets of good and bad architecture, and RL'ing that?
That is a question I have been asking myself. It was partially answered with Fable 5.1 and Astra. I worked with them on architectural design documents, and they were very good. Some of their suggestions were better than my solutions. They are not at a point where they can fully replace a system architect yet, but they're getting close.
There is still the open question of what to build and how to create a user interface that is understandable and presents complex information in ways understandable to humans. That's something AI cannot do yet.
Models have an alignment problem. As in, they are too aligned to push back hard enough on bad ideas. Sometimes you can’t really push back because of power dynamics and whatnot, but the job still needs to be done _somehow_. So humans then navigate it by sometimes just ignoring the bad ideas, sometimes playing along with the narrative, but then doing what _actually_ needs to be done, and sometimes, when there is no other way, digging in heels and killing some bad ideas before they’ve really taken root.
An agent could do this with the SOTA model capability that existed yesterday. But then they’d need to become seriously misaligned.
But when a human does that, would you say that the human is misaligned? Or are they really aligned, just on a deeper level?
The problem with AI alignment is that they are only superficially aligned. You can see this difference in people, too. Between the people-pleaser at the office, and the one who says hard truths that really need to be said (but that nobody else is willing to say), which one is really aligned?
I'm no senior but I think that if software (and rest of disciplines possible soon to be hit as hard) can still be "wrong", if you can still mess it up even with machine assistance, people doing software still have a place. This doesn't mean that we need to achieve absolute perfection to be deprecated, maybe we just become accepting of machine generated good enough artifacts. I'm betting on humanity infinite nonconformism instincts so I can have a job some more years!
>and rest of disciplines possible soon to be hit as hard
Other disciplines have way less publicly available training data than SW.
In other disciplines, the secret sauce of whether a technical decision is good or bad lise within the institutional knowledge and decades of experience of gray beards, not a compiler output that can be integrated in a machine's feedback loop.
Every vibe coder here that is bullish on AI needs to realize that we're training the AI. Helping it to replace us.
We're basically turkeys voting for Christmas.
We're cooked.
Being anti AI doesn't keep your job around for longer though.
I've started looking into creating a pest control business, or maybe renovating furnitures I find on the street. Let me know if you have other ideas for senior engineers.
I would love more existing or recently ex- software engineers to share what new activities they have been taking up for living and that have been going well. I am certain that there are niches where a refined systematic thinking that many software engineers have could be a big advantage. However, it's expensive and risky to start new businesses in areas we don't have any experience in.
FWIW - I think the small business market is actually very suitable for former SWE's. The capacity to troubleshoot processes, variance, bottlenecks, etc. is what turns a mom& pop shop into a regional chain.
I think it is the same category such as mathematicians, doctors, biologists, scientists, sales representatives, marketers, teachers, artists, influencers (yeah, these human influencers) and so on. Of course you can get rid off programmers or mathematicians but then you need to understand and handle stuff on your own. You still need somebody who can check "what's going on". Imagine you are boss. Do you really want to do all things on your own with agents? Sure, let's do it. I think some people can do it but there is a lot of people who prefer somebody else does it for them. You are boss, you know.
If you're building a business from the ground up, a new business, it's financially a better decision to build it as much as possible on agents, not people. It lowers your cost. And that's what people are doing because it's a new business and they don't have money yet.
Your competition, currently operating with high cost humans, will be forced to reduce their costs or fold.
So, if it's possible to run with fewer people, competition pressures everyone to do the same. You won't really be able to prefer humans soon.
The economy will dramatically change under unprecedented supply. Hopefully this happens slowly enough that we figure out what to do about it.
Ultimately you still need somebody to decide what we are going to tell the machines what to do, and then actually tell the machines what to do, and it doesn't really matter if that is in JavaScript or plain English.
There does _seem_ to be some pressure on the top end of SWE salaries. Its possible that this is just wishful thinking, but if true, I suspect that it has more to do with an increase in job applicants rather than a decrease in actual work. AI seems to have driven up the amount of applications received for every vacant position.
Disclaimer; I often feel dread around my career fading and needing to switch to a field I don't like & won't be paid well in.
That said, I think the rational perspective is "things will be fine, basically".
I tried writing a longer reply but organized thought drifted into ramblings about economics I only partially understand. TL;DR I think some opportunities will close and others will open, and current engineers will find a home. Mostly software architecture or as technical advisors for product companies (if everyone writes code, you need some Dude Who Knows It Well to guide the non-engineers when they get stuck. Akin to IT).
I think wages will stagnate but wages are generally inflated, if we're being honest. As much as I love my doctor-level salary, I can honestly say I don't deserve it.
almost officially graybeard territory, thinking about starting an apiary. getting stung by 100s of bees still sounds like more fun than being forced to have to interact with LLM code from any source
let me know when you're start recruiting
LLMs have been able to make good success in coding, because someone is reviewing the work it produces (code). Ofcourse quality of code has become better over time. IMO other than Human-in-the-loop scenarios, AI will not see huge adoption in businesses.
I don't think code quality has improved. I think the main advantage of LLMs is that they have unlimited perseverance to rummage in whatever code you give them to to solve a particular problem, and given enough loops they will also find and fix some of the bugs that have been either preexisting or introduced freshly. Code quality is important for us, humans, because a lot of times we don't have mental capacity, patience and speed to work in bad code bases for long time.
I moved out of SW Dev and into systems consultancy. Main moat of the (consultancy) job is that a customer will always need a human person to put the blame on. That's me. I guess similar roles are relatively safe.
Coding by hand will be prohibited. Too big of a risk factor. You may be the shaman communicating with the almighty.
I always fancied the title of Tech-Priest.
I am predicting a sharp rise in poets and painters.
Actually, the poetry thing is probably doomed too.
So maybe.. interpretive dance?
Tech is fucked and this site is mostly filled with an overly vocal minority of well compensated engineers who don't care.
Quality was already poor before AI. And now we can do it faster!
I second that
I think the idea of "software architecture" may become irrelevant.
Architecture, in the sense of applying/enforcing organising principles like modularity and levels of abstraction, is something we currently do largely to manage complexity -- so that we can fit as much of the system into our limited brains as we can. For a large enough system, this is necessary in order to work on it efficiently. But if you can "keep the whole thing in your head" without using these "crutches", then there's actually no need for them. They become overhead that can be done away with. We do this already for sufficiently small things -- no one plans a software architecture for a "Hello world" script.
It's possible that future (maybe even current?) AIs will be able to "keep the whole codebase in their heads" just fine, in which case the need for architecture goes away.
I don't disagree with your points, but there are several advantages to modularity that result in lower token use. I think there's $ in good architecture.
Once the Mother App to assign ubi tasks and assign ubi dorms and take orders for ubi gruel is complete, well that’s the only app we’ll need. May we all have pleasant ubi tasks to fulfill our days.
The only reason the tech bros are touting utopia is because they won’t be paying for it. It’ll be a tax on their customers, not them. This is a strange timeline.
This doesn't make any sense, unless you are just building low critical software such as games that assumes little to no liability.
What will happen is you will get less developers (including less seniors and staff-level) but there will always be expert humans that will check the code before release.
Maybe it was a good run for some here, but you need to look outside of the tech bubble. Do you want an LLM to autonomously fly a commercial plane without any pilots in 2030?
"Seniors" should already know this, but are quickly startled because they are unable to adapt since they based their entire identity and careers around the over-engineering of "clean code", "test driven development" and "architecture" too early on products that make no money, when the LLMs have now cheapened that and taken that role. Even Uncle Bob does not read the code Claude writes in 2026.
Instead, think like an owner.
I am a senior engineer. I adapted faster than most of the people I work with. It doesn’t matter that the tool can improve your productivity if you don’t know where to start which a lot of juniors don’t. They still need leaders to guide them and apportion the work.
AI still isn’t a magic bullet either. I spend some of my days undoing the unholy mess that generated code is.
> I think (businesswise) the need for actual software programmer skills is 90% out in a few years I think, and product lead etc people will be favored more as they have also more people skills.
"Instead, think like an owner." exactly, the traditional software engineer role has ceased to exist and now you must think like an indie entrepreneur and own things. There will be less than 10% of current engineers checking the code in the near future.
> Do you want an LLM to autonomously fly a commercial plane without any pilots in 2030?
No but I also didn't want LLMs to make decisions for the defense department in USA yet here we are.
> "Instead, think like an owner." exactly, the traditional software engineer role has ceased to exist and now you must think like an indie entrepreneur and own things. There will be less than 10% of current engineers checking the code in the near future.
It will still exist, but far less of them and it includes the reduction of SWEs from junior to staff-level.
Someone will always be held liable if an LLM runs autonomously and out of control which will cost them a lot of money for damages.
> No but I also didn't want LLMs to make decisions for the defense department in USA yet here we are.
That was someone else (another human) that made that decision to use LLMs there, and why certain countries don't trust the US any more, which is my point with both liability and trust in highly critical situations.
I think we just both agree on our views? I don't think all software engineering is going to die and especially those who can be held liable can keep their jobs, but otherwise I think the profession is commodified so much that anyone can do it in 5 years - e.g. "Claude tell me a prompt to use to make this software faster and less buggy"
Sigh. I never want to be a product people or an owner. Fortunately I’m over 40 and don’t have to work for very long. Good luck folks. I’ll resort to recreational programming when I half retire.
I'm almost 40, but nowhere close to being able to retire and enjoy the rest of my life, which is, I hope, going to be at least another 40 years. It's going to be a fun ride
I wish you good luck. On my side I actually don’t have the amount of money to retire comfortably. I’ll just retire uncomfortably, like going back to the younger days when I spent 1000 bucks a month for everything.
I mean, you do own your work in a way... it's not so different.
Yeah…but it’s a completely different mindset. I’m not saying mine is better —- actually probably worse, but I’m who I am.
>Instead, think like an owner.
Exactly. LLMs make it, I'd say, _possible_, to own way more parts of the code than before, and understand even more.
> Do you want an LLM to autonomously fly a commercial plane without any pilots in 2030?
This argument has been made before LLMs and it didn't make sense to me then either. Sure I do if it crashes statistically less often than humans!
But sure, if my job now is going to jail when something malfunctions, that's still better than seeing my family starve.
why do you believe what you think matters?
Just want to stir up some discussion, get other viewpoints in from humans. Nothing more than that
Next week I will power on a physical server in a colocation. The server is a vertical scale running my SaaS that I developed over the years and needs more space now. Sure, LLMs help me now with a lot of things. Coding, server commissioning, marketing, etc., but I simply cannot image how the LLM drives my car to the colocation and plugs the server in?!? How should this work? And who is responsible at all in the end? Responsibility? And renting something "automatic" like AWS would cost 20-30 times more btw. This is not an option.
Well, a colo will typically offer a Remote Hands service for $/mo and do that part for you, but its a human.
I suppose eventually it could be a rack robot instead, why not.
An LLM can order a service to do it for you.
My wallet can't.
Why can't a LLM design and run a robot?
That drives my car, orders the parts and builds the server?
Perhaps you need 3 robots? All of those things can be done. It takes some effort, so it will take a while, but robots have been making inroads since the spinning Jenny and power loom - both from the 1700s.
Self driving cars already work well in niche (millions of people have used them successfully). There is debate over how well (the only people with access to real world data who are talking also have a clear bias so who knows), but they at least work today.
There are a lot of order systems that automatically order parts for you. Just in time and tracking what you ship makes it easy for a program to figure out what you need. If you order something from my company the system automatically orders all the parts needed to arrive the day we assembly in our factory (which is very automated and getting more automated all the time).
I'm sure places like Dell have a lot of automation in their build a server, though I don't have insight into how they do this. Likewise every medium sized company has images for everything so all the software can be automatically installed.
None of the above is perfectly human free yet. However it has been moving that way for years and will continue to. LLMs have a place in moving that long.