I'm ~20 years into my career, currently working on a relatively complex C++ codebase for a space application. Internally we have copilot (gpt 4 or claude 3.7) and a custom ChatGPT (4o/o3) client we can use. It's great for autocomplete, boilerplate stuff, and generally learning about a new topic but terrible for complicated or niche development (just outright fabricates stuff that doesn't exist or compile).
We got a batch (10+) of summer interns this summer, bright students from good schools, and they are absolutely glued to the AI tools. They're getting lots of code written with lots of tasks complete, but occasionally they write code that makes absolutely no sense and can't explain why they did it. Lots of very overcomplicated solutions to problems, and seemingly low comprehension of the code they're creating. It actually makes me feel pretty good about career viability long term -- people that have a deep understanding of the systems and code will need to be around to clean up the mess that is about to be created over the next 5-10 years.
I don’t see AI as a threat to programmers now or in the future — it’s just a tool. Think of programming like cutting down trees: we used to swing axes, and now we’ve got chainsaws. But even the best chainsaw won’t cut a tree on its own — it still needs the woodcutter.
AI is trained on what humans have already done. It can remix and automate, sure, but it doesn’t innovate or create like humans do — at least not yet. Reaching that level of creativity is a whole different game, and if AI ever does get there, we’ll have way bigger questions to worry about than job automation.
Bottom line: AI won’t replace us — it’ll reshape how we work. The role of the programmer will evolve, just like it has with every major shift in tools and tech.
Download claude code. Use it. Try to build that thing that would take 5 people six months to do. You'll find very quickly that programmers are needed, and more skilled ones too.
For a long time, we've been coders, developers. Mentally in the code. But now we work on the code. We're now engineers. We coordinate how the parts work together. I like this definition of engine: "any of various mechanical appliances, often used in combination"
The most expensive model today is Claude Opus 4. It can draft plans. It can engage this plan. It can write tests, run tests, fix tests. All by itself. But if you don't keep an eye on it, it will run off and do something you didn't intend it to do for 20 minutes.
In the long term, it depends on how good AI systems get, but right now they have a long way to go before they can be trusted to work autonomously. Junior devs are an investment, and even before AI were a net negative to productivity; they’ve always been hired for their potential and capability to grow. AI changes nothing here. The biggest danger I see is if AI disrupts the mastery process somehow for these devs.
In the short term, naive bosses may try to offload too much work to AI, but someday they’ll realize their mistake. I don’t think programmers are going anywhere.
As a very seasoned developer, I have seen lots of projects built by a single person in a short time that “would normally” be built with more people over a longer period of time. What is invariably missing from these projects are the very things that the 5 people would have added. The main thing that is usually missing is robustness. These single dev projects will usually fall over hard when they stray from the happy path (why would anyone enter a value other than a small digit for this input???).
Data integrity is often ignored as well, both internal to the application and how the data flows through the rest of the systems and the org as a whole.
Security is generally an afterthought as well.
And then, there’s sort of the elephant in the room, which is part of why these projects take a long time, is that they need to be defined and refined in the first place. It is not uncommon to find some edge case that requires a lot of effort to decide how best to handle it, and it’s often the product owner who needs to decide tradeoffs for thw product.
Incidently many of these reasons are why offshoring also fails.
So, where does that leave things? AI will undoubtedly allow a team of 9 to do what a team of 10 did 2 years ago (maybe even 3 of 5). AI also will potentially help with each of the things I’ve mentioned as well, I’ve had good luck with AI helping with missing test coverage. I’ve also had a lot of luck with AI helping where I have gaps in my knowledge about things. So I expect the quality of software will get much better too.
I will say, right now there is a big shake up of the industry. I expect short term prospects to be bad in the same way the offshoring frenzy of the early 2000s. What you are going to find, is that long term demand for technical talent will only go up. In the same way that efficiency in anything tends to increase demand, things get cheaper and people find more uses for said thing.
Another thing to think about; I sort of saw the writing on the wall 6 years ago. I continually saw the barrier to entry fall in software. So I changed my niche and moved closer to the hardware, where fewer people seemed to be entering. I will tell you that the places LLMs struggle with the most are the places others haven’t spent much time documenting. In the domain I currently work, it is impossible to use any code auto generation, there are too many places it transposes things or makes up api calls. But the code it generates can be very useful when the documentation is ambiguous and some obscure forums post likely has an answer the LLM can draw from. So it is possible to stay ahead of AI by moving to the fringes.
If you want to stay as a software developer. Lean into AI. As per bill gates. Lean into AI Engineering. Develop the models. This will be an in demand skill for 100 years.
Get that AI Engineering master's degree. Learn the math, the statistics, the models.
Right now it’s fairly limited , hence the many ‘programmers will always be needed’ opinions.
What worries me is the trend and speed at which things are progressing, especially with the amount of money thrown at it.
If you believe the trend will stay then many of us will be out of a job in a few years. If you don’t then you’re safe.
There are already markets ( translation, illustration etc ) where AI has changed things significantly. Ours is slightly different because it requires formal things to be correct unlike translations or illustrations which can be slightly ‘off’.
My view is that either AIs learn how to generate human debuggable or already correct code, or it won’t work and the trend will stop. I believe there’s a fundamental hardness here - if they break that , then other things will change and our world will be fundamentally different.
AI is an accelerator if you know one language and you're trying to go full-stack, say. The learning you get is incredible. All other use-cases are dubious. If you're a senior dev, you're still faster (once you include the debugging). If you're a junior dev it's too buggy and gets out of control as complexity increases.
I think AI is good only for assisting with repetitive tasks, generating boilerplate code, and even suggesting optimizations. As AI improves, the demand for developers who can design workflows, plan software architecture, and address security issues will grow. For example, I just started using the replit to make a website but I decide the structure, flow, and security of the website. I handed over repetitive tasks to QA guys. So in my case, it is helping me and generating more revenue. May be after a decade even a non tech guy could develop it completely but for now this is not the case.
I would look at it from a demand-supply perspective. The demand will significantly reduce, and the supply has increased. I also love to look at it from a survival of the fittest perspective. If you are genuinely good at what you do and can drive exceptional results, you don''t have to worry. But if not, you might have to find something where you can "actually provide value" and not just be a fly on the wall.
not convinced it will have a huge impact, the reason being is - if it would be possible to automate the job of software developer 100%. you would in effect automate all and any white collar jobs overnight.
Honestly? I have no idea. It's too soon to tell, really. We have to wait until the hype cloud starts to dissipate and we can see what the land actually looks like.
> This made me worry more than ever for the threat of becoming obsolete on my career due to the evolution of AI. What do you all think here?
Whatever impact genAI has, there will be a strong demand for traditional devs for years to come. Worst case, where that demand comes from may change (for instance, it might not be from SV style companies), but it will still exist.
I think, given the frequency of AI posts, and related slop on HN, and the statistically significant number of posts showing de-amplification happening on anyone responding to AI sentiment critically on HN (within 3-10 hours of the post) means a few things.
Anyone responding is going to find themselves downvoted (karma leached) regardless of any legitimate points or position and you can't have interesting conversations without measured disagreement. Value comes from things you don't know, "yes people" blindly follow the trend.
There's also no real point or value in a discussion when you have bad actor bullies silencing truth they disagree with or opinions that are rationally based, and then have them trying to gaslight and play off the lack of anything to the contrary as their sentiment is just what everyone thinks (because downvoted posts disappear beyond a threshold).
Its not, and HN is just a minor step up from 4chan with perspectives favoring Chinese talking points most times, almost completely unmoderated with few exception. The sophisticated propaganda, manipulation, and other stuff just flows freely.
Just recently I had a post where I was significantly downvoted for saying AI shouldn't be controlling nuclear power plants for safety issues. How's that for warped. The people were saying the AI could manage the power grid and just turn the power plant back on right after it was shut down, and I pointed out there's a 3-day wait and failing that wait is what happened at Chernobyl and you can't do that in a controllable way (Xenon is a problem). -9 on that post, and flagged; now its gone afaics.
Seriously messed up. There's a point where the platform gets a bad reputation, and crazy/delusional polarizing elements take over by driving the grounded and measured people away.
I'm ~20 years into my career, currently working on a relatively complex C++ codebase for a space application. Internally we have copilot (gpt 4 or claude 3.7) and a custom ChatGPT (4o/o3) client we can use. It's great for autocomplete, boilerplate stuff, and generally learning about a new topic but terrible for complicated or niche development (just outright fabricates stuff that doesn't exist or compile).
We got a batch (10+) of summer interns this summer, bright students from good schools, and they are absolutely glued to the AI tools. They're getting lots of code written with lots of tasks complete, but occasionally they write code that makes absolutely no sense and can't explain why they did it. Lots of very overcomplicated solutions to problems, and seemingly low comprehension of the code they're creating. It actually makes me feel pretty good about career viability long term -- people that have a deep understanding of the systems and code will need to be around to clean up the mess that is about to be created over the next 5-10 years.
I don’t see AI as a threat to programmers now or in the future — it’s just a tool. Think of programming like cutting down trees: we used to swing axes, and now we’ve got chainsaws. But even the best chainsaw won’t cut a tree on its own — it still needs the woodcutter.
AI is trained on what humans have already done. It can remix and automate, sure, but it doesn’t innovate or create like humans do — at least not yet. Reaching that level of creativity is a whole different game, and if AI ever does get there, we’ll have way bigger questions to worry about than job automation.
Bottom line: AI won’t replace us — it’ll reshape how we work. The role of the programmer will evolve, just like it has with every major shift in tools and tech.
Download claude code. Use it. Try to build that thing that would take 5 people six months to do. You'll find very quickly that programmers are needed, and more skilled ones too.
For a long time, we've been coders, developers. Mentally in the code. But now we work on the code. We're now engineers. We coordinate how the parts work together. I like this definition of engine: "any of various mechanical appliances, often used in combination"
The most expensive model today is Claude Opus 4. It can draft plans. It can engage this plan. It can write tests, run tests, fix tests. All by itself. But if you don't keep an eye on it, it will run off and do something you didn't intend it to do for 20 minutes.
In the long term, it depends on how good AI systems get, but right now they have a long way to go before they can be trusted to work autonomously. Junior devs are an investment, and even before AI were a net negative to productivity; they’ve always been hired for their potential and capability to grow. AI changes nothing here. The biggest danger I see is if AI disrupts the mastery process somehow for these devs.
In the short term, naive bosses may try to offload too much work to AI, but someday they’ll realize their mistake. I don’t think programmers are going anywhere.
>5 people over a span of 6 months
As a very seasoned developer, I have seen lots of projects built by a single person in a short time that “would normally” be built with more people over a longer period of time. What is invariably missing from these projects are the very things that the 5 people would have added. The main thing that is usually missing is robustness. These single dev projects will usually fall over hard when they stray from the happy path (why would anyone enter a value other than a small digit for this input???).
Data integrity is often ignored as well, both internal to the application and how the data flows through the rest of the systems and the org as a whole.
Security is generally an afterthought as well.
And then, there’s sort of the elephant in the room, which is part of why these projects take a long time, is that they need to be defined and refined in the first place. It is not uncommon to find some edge case that requires a lot of effort to decide how best to handle it, and it’s often the product owner who needs to decide tradeoffs for thw product.
Incidently many of these reasons are why offshoring also fails.
So, where does that leave things? AI will undoubtedly allow a team of 9 to do what a team of 10 did 2 years ago (maybe even 3 of 5). AI also will potentially help with each of the things I’ve mentioned as well, I’ve had good luck with AI helping with missing test coverage. I’ve also had a lot of luck with AI helping where I have gaps in my knowledge about things. So I expect the quality of software will get much better too.
I will say, right now there is a big shake up of the industry. I expect short term prospects to be bad in the same way the offshoring frenzy of the early 2000s. What you are going to find, is that long term demand for technical talent will only go up. In the same way that efficiency in anything tends to increase demand, things get cheaper and people find more uses for said thing.
Another thing to think about; I sort of saw the writing on the wall 6 years ago. I continually saw the barrier to entry fall in software. So I changed my niche and moved closer to the hardware, where fewer people seemed to be entering. I will tell you that the places LLMs struggle with the most are the places others haven’t spent much time documenting. In the domain I currently work, it is impossible to use any code auto generation, there are too many places it transposes things or makes up api calls. But the code it generates can be very useful when the documentation is ambiguous and some obscure forums post likely has an answer the LLM can draw from. So it is possible to stay ahead of AI by moving to the fringes.
No kidding. There's a lot more to software development than mindless coding.
If you want to stay as a software developer. Lean into AI. As per bill gates. Lean into AI Engineering. Develop the models. This will be an in demand skill for 100 years.
Get that AI Engineering master's degree. Learn the math, the statistics, the models.
https://www.indiatoday.in/technology/news/story/bill-gates-s...
Right now it’s fairly limited , hence the many ‘programmers will always be needed’ opinions.
What worries me is the trend and speed at which things are progressing, especially with the amount of money thrown at it.
If you believe the trend will stay then many of us will be out of a job in a few years. If you don’t then you’re safe.
There are already markets ( translation, illustration etc ) where AI has changed things significantly. Ours is slightly different because it requires formal things to be correct unlike translations or illustrations which can be slightly ‘off’.
My view is that either AIs learn how to generate human debuggable or already correct code, or it won’t work and the trend will stop. I believe there’s a fundamental hardness here - if they break that , then other things will change and our world will be fundamentally different.
Note, this is all belief !
AI is an accelerator if you know one language and you're trying to go full-stack, say. The learning you get is incredible. All other use-cases are dubious. If you're a senior dev, you're still faster (once you include the debugging). If you're a junior dev it's too buggy and gets out of control as complexity increases.
I think AI is good only for assisting with repetitive tasks, generating boilerplate code, and even suggesting optimizations. As AI improves, the demand for developers who can design workflows, plan software architecture, and address security issues will grow. For example, I just started using the replit to make a website but I decide the structure, flow, and security of the website. I handed over repetitive tasks to QA guys. So in my case, it is helping me and generating more revenue. May be after a decade even a non tech guy could develop it completely but for now this is not the case.
Just watched a CNBC article on you tube: 'Rethinking the AI coding payoff'
https://www.youtube.com/watch?v=shOLHtyUaFg
Basically it augments or provides boiler plate but that is all.
I would look at it from a demand-supply perspective. The demand will significantly reduce, and the supply has increased. I also love to look at it from a survival of the fittest perspective. If you are genuinely good at what you do and can drive exceptional results, you don''t have to worry. But if not, you might have to find something where you can "actually provide value" and not just be a fly on the wall.
not convinced it will have a huge impact, the reason being is - if it would be possible to automate the job of software developer 100%. you would in effect automate all and any white collar jobs overnight.
Honestly? I have no idea. It's too soon to tell, really. We have to wait until the hype cloud starts to dissipate and we can see what the land actually looks like.
> This made me worry more than ever for the threat of becoming obsolete on my career due to the evolution of AI. What do you all think here?
Whatever impact genAI has, there will be a strong demand for traditional devs for years to come. Worst case, where that demand comes from may change (for instance, it might not be from SV style companies), but it will still exist.
I am in denial.
This sounds a lot like acceptance.
I think, given the frequency of AI posts, and related slop on HN, and the statistically significant number of posts showing de-amplification happening on anyone responding to AI sentiment critically on HN (within 3-10 hours of the post) means a few things.
Anyone responding is going to find themselves downvoted (karma leached) regardless of any legitimate points or position and you can't have interesting conversations without measured disagreement. Value comes from things you don't know, "yes people" blindly follow the trend.
There's also no real point or value in a discussion when you have bad actor bullies silencing truth they disagree with or opinions that are rationally based, and then have them trying to gaslight and play off the lack of anything to the contrary as their sentiment is just what everyone thinks (because downvoted posts disappear beyond a threshold).
Its not, and HN is just a minor step up from 4chan with perspectives favoring Chinese talking points most times, almost completely unmoderated with few exception. The sophisticated propaganda, manipulation, and other stuff just flows freely.
Just recently I had a post where I was significantly downvoted for saying AI shouldn't be controlling nuclear power plants for safety issues. How's that for warped. The people were saying the AI could manage the power grid and just turn the power plant back on right after it was shut down, and I pointed out there's a 3-day wait and failing that wait is what happened at Chernobyl and you can't do that in a controllable way (Xenon is a problem). -9 on that post, and flagged; now its gone afaics.
Seriously messed up. There's a point where the platform gets a bad reputation, and crazy/delusional polarizing elements take over by driving the grounded and measured people away.
That topic is a much more interesting topic.
[dead]