I think we’ll adapt to whatever AI throws our way. Think about it when Excel came along, it took away a lot of the boring, repetitive stuff in accounting. The accountants who learned to use it didn’t lose out they actually did better. That’s just how technology works.
In IT, change has always been part of the job. You have to keep learning and adjusting. That’s nothing new. I don’t think that’s going to change anytime soon. Five years from now, the people who stay curious and flexible will still be the ones getting the best jobs. It’s always been that way, and it probably always will be.
One of the issues with the current crop of CS graduates is that they were sold on the idea that they could get a job once they graduate. But they never got the idea that IT changes fast. So part of the job is to always be learning new stuff to keep a job and get the next job. It's a hard career that way. So we'll see lots of CS graduates that could not keep up and are jobless or moved on to other fields.
Nobody can predict the future about this sort of thing with any kind of accuracy, but I'll throw my guess out there anyway. For reference, I'm a software engineer with 12 years of experience at a few companies big and small.
The AI bubble will have burst, and the market will have started to heal from the AI enshittification in some areas but still be grappling with it in others. The "you must use AI" mandates will start to get outnumbered by "you may not use AI" mandates. AI will have gotten only a little better (mostly just better tooling and cheaper operating costs, but it will be more expensive), still nowhere near being able to replace actual humans doing skilled technical work. LLMs will have found product market fit in the NSFW market and in the online scammer market.
There will be a glut of senior engineers and a resurgence in demand for developers for a few reasons:
- The juniors who fail to find jobs now, will go into other professons and will not become senior engineers.
- AI slop will have creeped into, and caused mayhem in, many large and important codebases, and it will have introduced some major security vulnerabilities which will have produced at least a few major news events. Companies will be forced to hire senior engineers to deal with this slop and the massive technical debt that was created by LLMs.
The increased demand for senior engineers will likely be met largely by further mass importing of H1Bs, and outsourcing, not as much domestic hiring. Many kids from the generation of CS majors who are graduating now (the ones everyone told to "learn to code" for their whole childhood) will have failed to find employment in software engineering jobs, "because of AI," (the other "AI," iykyk). They will blame this on Americans not being willing to study STEM, being too lazy, etc. They will conveniently forget how Americans did study STEM, they did learn to code, and when they graduated, they were met with a job market that was shit because the AI grifters duped business guys into not investing in the next generation.
The generation coming up behind them has been told that all jobs will be eradicated by the time they graduate, so yeah, you can imagine what that's doing to the minds of our kids.
Or one could be pessimistic, AI code remains shite but "good enough" to employ former devs as dehallucination operatives, grinding through and fixing up code which no-one understands as best they can for just enough money to keep scurvy at bay.
I think the bar will raise a little. I want to see a little more from new hire devs that are beyond coding exercises, because Copilot has solved most of those.
I think we’ll adapt to whatever AI throws our way. Think about it when Excel came along, it took away a lot of the boring, repetitive stuff in accounting. The accountants who learned to use it didn’t lose out they actually did better. That’s just how technology works.
In IT, change has always been part of the job. You have to keep learning and adjusting. That’s nothing new. I don’t think that’s going to change anytime soon. Five years from now, the people who stay curious and flexible will still be the ones getting the best jobs. It’s always been that way, and it probably always will be.
One of the issues with the current crop of CS graduates is that they were sold on the idea that they could get a job once they graduate. But they never got the idea that IT changes fast. So part of the job is to always be learning new stuff to keep a job and get the next job. It's a hard career that way. So we'll see lots of CS graduates that could not keep up and are jobless or moved on to other fields.
Nobody can predict the future about this sort of thing with any kind of accuracy, but I'll throw my guess out there anyway. For reference, I'm a software engineer with 12 years of experience at a few companies big and small.
The AI bubble will have burst, and the market will have started to heal from the AI enshittification in some areas but still be grappling with it in others. The "you must use AI" mandates will start to get outnumbered by "you may not use AI" mandates. AI will have gotten only a little better (mostly just better tooling and cheaper operating costs, but it will be more expensive), still nowhere near being able to replace actual humans doing skilled technical work. LLMs will have found product market fit in the NSFW market and in the online scammer market.
There will be a glut of senior engineers and a resurgence in demand for developers for a few reasons:
- The juniors who fail to find jobs now, will go into other professons and will not become senior engineers.
- AI slop will have creeped into, and caused mayhem in, many large and important codebases, and it will have introduced some major security vulnerabilities which will have produced at least a few major news events. Companies will be forced to hire senior engineers to deal with this slop and the massive technical debt that was created by LLMs.
The increased demand for senior engineers will likely be met largely by further mass importing of H1Bs, and outsourcing, not as much domestic hiring. Many kids from the generation of CS majors who are graduating now (the ones everyone told to "learn to code" for their whole childhood) will have failed to find employment in software engineering jobs, "because of AI," (the other "AI," iykyk). They will blame this on Americans not being willing to study STEM, being too lazy, etc. They will conveniently forget how Americans did study STEM, they did learn to code, and when they graduated, they were met with a job market that was shit because the AI grifters duped business guys into not investing in the next generation.
The generation coming up behind them has been told that all jobs will be eradicated by the time they graduate, so yeah, you can imagine what that's doing to the minds of our kids.
Or one could be pessimistic, AI code remains shite but "good enough" to employ former devs as dehallucination operatives, grinding through and fixing up code which no-one understands as best they can for just enough money to keep scurvy at bay.
I think the bar will raise a little. I want to see a little more from new hire devs that are beyond coding exercises, because Copilot has solved most of those.
this is far too open ended and too far into the future for anyone to give a response that is not deeply speculative
Probably not that hard.