How AI is upending the software development industry
reuters.com
reuters.com
The idea is likely flooding the market with devs, lowering software development costs and increasing software in the market increasing cloud usage.
If I were to put on my tinfoil hat though, I agree with your theory. These people often (in my personal experience) don’t know how to do anything not covered explicitly in the learning or testing material, leading to outsized dependencies on subscription or vendor portfolio products in lieu of independent or collaborative problem solving.
I've always felt like coding bootcamps were pretty much doomed to fail. I don't know why so many in the industry tried to denigrate the profession in the first place by pretending that a ~20 week bootcamp would suffice. It's like if the New York Philharmonic said "We're going to expand beyond the old stuffy prestige names of Julliard and Curtis, and instead start looking for non-traditional students who took a semester long 'Learn the Cello in 12 weeks' course."
But even then, the biggest problem with coding boot camps wasn't competition from AI, it was offshore competition. People in the US spent a ton to be trained in basic skills like front end web development when there are millions of people around the world willing to do that, at a similar or higher level of skill, for $20/hr or less.
I don't mean to denigrate people who took non-traditional routes to software development. On the contrary, some of the best programmers (and, especially, dev managers) I've ever worked with don't have CS degrees. Speaking of cello, for some reason an inordinate number of top programmers I've worked with have music degrees. But all of these folks just had a certain way of thinking - very detail oriented, extremely logical, extremely self-motivated, that is just not common among people at large. It's not surprising companies try to find talent where the pool is full rather than doing needle-in-a-haystack searches.
Also, there are unforeseen consequences. Take venture capital, its actually relatively easy for a harvard grad to raise money, the reason being:
- everyone is trying to get an early piece of a company from the next bill gates/mark zuckerberg
- they fund every harvard grad since returns on a win like that are 10,000X.
-> consequence is almost all early stage capital gets allocated to ivy grads
-> most companies owned by ivy grads
-> caste system
the hope is that ai eliminates this vetting process by giving leverage to everyone
I agree, but then again, if you look at areas like sports and the arts, the vast majority of people who are stars were also standouts when they were 14 (or often younger). It doesn't do any good to pretend that someone who was coding since elementary school wouldn't have a huge (and warranted) advantage over someone who didn't pick it up until their 20s.
I also agree people get into top colleges for lots of different reasons, but (at least in my experience), graduating from a top school just got your foot in the door, but you still had to go through the same judgement gauntlet as everyone else after that.
1. legacy admits
2. athletic admits
3. dei admits
4. actual brilliances admits
but even in the 4th bucket, they are allocating those spots based on alot of non performative criteria.
with sports, its more like, "who ran 40 meters the fastest"
I live in NYC, I meet the ivy types all the time. I have gotten very cynical about it
I was accepted to MIT, but my family wasn't poor enough to receive need-based aid, and we certainly weren't wealthy enough for me to go without aid (especially because it would have been out-of-state tuition), so I ultimately went elsewhere. I expect a fair number of others here have similar stories.
A big part of the reason ivy-league and other elite schools are selected for is that they are a strong signal for family wealth.
I'm calling bullshit on this story because MIT is not a public school - there is no in-state vs. out-of-state tuition, and as someone who did a ton of cost comparisons on the private vs. public schools where I was accepted, I have difficulty believing someone who was accepted to MIT would confuse this fact.
What I do remember was that MIT was so incredibly expensive that I didn't have to give it much thought -- it was never a viable option due to cost and lack of financial aid possibilities.
Boot camp on its own though.. I don't disagree with you there. I was never convinced this stuff could be properly learned in such a short period of time. Perhaps just well enough to pass an interview. Combine that with some self learning and freelance work, then quite sure.
That's the statement I disagree with. Software companies (good ones anyway) have plenty of tools to vet candidates. It's just that they'd rather go fishing where the pool is full than where it's mostly minnows.
Arguably, for the type of developer this article is discussing, I see AI coming for the junior “mathematical” or “algo” developer before the product/feature focused developer.
Besides morals, you'll always need people in the interface between the computers and the world. Maybe they won't write too much code, but they'll need to specify and verify behaviours anyway.
With the rise of AI it will only get worse.
Not sure if there is a single silver bullet to solve this, but one thing is all juniors could start building things only for fun and solving small problems they understand (ala YC advice), and once solved, allow room to build it better each time.
It actually doesn’t make sense when you think about it - the idea that entry level roles will be cleanly eliminated. Certainly there could be a period with drastically fewer entry level roles, but that logically can not remain the case.
Rolling the ladder up behind us - https://news.ycombinator.com/item?id=44328894 - June 2025 (201 comments)
Just like how CAD has almost entirely eliminated that line of work, the same thing will happen in software development. Only the “architects” will remain.
There is an implicit assumption made by the people concerned about junior devs disappearing that that’s where senior developers come from.
The reality is that almost junior developers never gain significantly better skill, no matter how their age or title.
I regularly work with people with decades of “experience” that still have no idea how to use a database engine. A real senior developer could write their own database engine from scratch. That’s the difference!
If they don’t know that, then the database engine internals will forever remain a mysterious black box to them.
Explaining each of the four letters in ACID is the follow-up question if they pass.
These are the equivalent of “why do we put steel rebar into concrete?” If you can’t answer these questions immediately upon graduating then you cannot ever be responsible for anything. You can only be entrusted to copy or “fill in the blanks” where an anctual engineer worked out the design.
I mean I guess they can choose to jump teams or jobs every couple years to gain some breadth, but it seems cruel to force people to hop around to move up the ladder. (Even though I admit this is still the fastest way, but for different reasons!)
Seniors will have to come from somewhere, but it wouldn't be through this pipeline, even if companies were still hiring at full rate for entry-level work.
For some just getting into the field, this could be a great opportunity -- if you can figure out where the puck is heading, you can skate directly there.
I've interviewed more that a few "software architects" who couldn't put together a for loop, and these folks were largely just BS artists and PowerPoint generators.
But yeah I'm sure increasing energy input into symbolic systems will not increase their entropy and thus the demand for workers capable of formal logic to keep it in check.
Because dumping representations of formal logic on a exponentially increasing repository of compressed data will surely start giving out output with higher informational content than the input. Any time now. Just double the size of the model a few more times and we will start getting lower entropy outcomes for free.
OK, so I think there is a fair chance it will play out like this: the younger generation won't study CS anymore / learn how to code. Instead, they'll all vibe code.
This will lead to a massive increase in demand for seniors to clean up the vibe code, which isn't working properly or needs maintenance.
This demand increase will be exacerbated by the fact that there will be no new seniors since no one learned how to code...
Fun times ahead.
Assuming you find cleaning up vibe (legacy) code fun haha
Before a big shift: “What could possibly go wrong?”
After a big shift and facing unintended consequences: “How were we supposed to know?”
Same after the dot com crash and outsourcing. Give it 5 years and hiring will be back to normal. Give it 10 and salaries will be even more stratospheric because the pipeline is bone dry now.
I'm not sure how bringing up outsourcing helps what you're saying, because tech salaries went ballistic shortly after outsourcing became popular.
Funny how that was also said at those times.
On a serious note, can't wait to see what salaries I command as a senior dev in a few years...when junior dev pipeline dried up and other older seniors left the industry.
More likely, they'll lay everyone off and with some luck start hiring them back at same or lower salaries and tell us we're lucky to even have a job, and we'll have no choice but to accept...
Those skills you don't fully develop at school or university, let alone a half year of bootcamp. I believe you have to have a developer's or engineering mindset in some way, early on in life, to be good at it.
My point being, if you choose a career swap, or suddenly realise you were studying something you are not passionate about, you still have a very big leap to make.
Sounds like it's the story of 1200 graduates, actually.
Don't forget that a software company is bunch of software engineers in coordination, if those software engineers can be placed with a graphics card or AI service provider they just go ahead and start your own software company and just use this services to undercut the software companies that no longer need software engineers.
If AI is actually able to code like a real software engineer then people that are not software engineers and were laid off from software companies should just get an AI subscription and take away the business from the company that they were fired from.
Maybe it's not that graduates are not able to get jobs because AI is replacing them but because the focus of the capital at this time is no longer on speculating over software projects, therefore on the companies that are actually making money are able to actually hire. Today if you want to work in a company that exist thanks to speculative capital and pays huge salaries like in the day when software was eating the world than you need to be in AI.
Leading up to the dotcom bust was a strong demand for technical talent. Talent that graduated in the midst of that bust got hosed, though were able to pick up work again as the economy realized that despite the bubble, technical expertise was still sorely needed.
Then came the 2000s, and the demand for technical talent skyrocketed on the back of Web 2.0 and the Digital Transformation of Business. Every college and grift was churning out the thinnest credentialed junior technologists to fill vacancies. Then the global recession happened, and suddenly every business started getting onboard the outsourcing wagon, thus crippling domestic technical roles and long-term IT careers. Except when ZIRP happened, suddenly everybody wanted software developers out of this notion that they could automate their way out of any conceivable problem with code.
Now LLMs are here, trained on the past and reducing the need for Juniors in those fields. The glut of C and D “bootcamp” students are superfluous as the business world re-orients around (comparatively) expensive debt and interest rates, and as technological revolutions effectively stagnate outside of a few, narrow fields (Quantum Computing, Cryptography, etc). AI coding tools built on LLMs are likely to decrease the roles of Juniors in orgs except for those high-quality candidates who work predominantly in newer languages or technologies LLMs can’t be effectively trained upon due to a substantially thinner corpus than, say, Javascript or C++.
Those of us who lived through prior cycles already see what’s coming next: refusing to hire Juniors ultimately cripples the pipeline of Senior talent, and educational institutions aren’t effectively training Seniors out the gate (since that requires both learned knowledge and lived experience). Seniors will likely appreciate in value in the short term (especially if LLMs prove my doomerism wrong and stick around/remain popular), followed by a business panic as they realize they footgunned themselves in the labor department by failing to hire, mentor, and train Juniors into Seniors. Combined with population growth declines, increasing retirement of older workers, and rising Nationalism clamping down on immigration or outsourcing, and you’ve got a perfect storm of sorts where companies will need talent but will have a lack of supply to pull from.
If I were a technologist whose background is bootcamps or vendor certifications, I’d be seriously worried about immediate and short-term prospects. Even if they hallucinate consistently and reward hack frequently, LLMs can supplant the kind of low-quality talent who normally take these roles in larger companies (at least for now). My advice is to start building more complex skill sets instead of relying on drills and rote memorization; contributing to FOSS isn’t enough when you’re competing against LLMs and thousands of randos for commit recognition, you’re going to have to make your own projects that solve problems or demonstrate your specific capabilities.
A and B-tier talent will have a rough few years as well, but they’re likely to soldier onward relatively unscathed unless I’m wrong and LLMs somehow lead to the Singularity. If you can’t get into that sort of quality, you shouldn’t pin your career hopes on a bootcamp or certification alone.