After 5+ years of actively trying to get into the field (pre AI), I left.
I threw my degree in the toilet, I closed my linkedin, and I went to go work in the trades as a diesel mechanic.
Greatest choice I've ever made. The pay is great, the work is steady, the coworkers are relaxed and not trying to one up each other. I'm now being paid to go to school, and get raises every year until I'm fully ticketed (way more than I ever made in the entry tech positions).
I've heard non-stop my whole life that if I join a trade it's going to be grunt labor and I'll be paralyzed in 5 years. Maybe some are, but this isn't hard at all. I lay on a creeper and turn wrenches. Anything over 50lbs we have lifts for.
Tech has become fun again, I'm just making projects because its what I wanted. I come home and relax by writing on my projects.
Now, I'm watching my tech friends from a distance and my only regret is not doing this sooner.
Reading this and some of your other comments in this thread, I think it’s awesome you’ve landed up doing what you really enjoy and are well compensated for it.
It makes me wonder if I would be happier doing something else, but (because of my personality) I’m very doubtful.
Since you see yourself as also being a computer guy I’m assuming that lack skill or intuition was not why you left the industry, so don’t read the rest of my comment as talking about you.
But I’ve definitely seen plenty of people in the software development industry where they may get by “okay” at their job, but things don’t tend to “click” as easily (in terms of intuitive understanding) for them the same way they do for me.
So I feel lucky and deeply happy to be at a company I enjoy working at and doing what has always been my passion.
It’s not that the computer industry is completely terrible (although plenty of parts of it certainly are), it’s just that for some people it’s not their true passion (which is fine).
You might be able to be a diesel mechanic and clean up in tech. This video is almost a year old, about applying Claude Code to boring businesses, a mobile diesel mechanic service in this case study: https://www.youtube.com/watch?v=gWNFna6fgS8
> I closed my linkedin, and I went to go work in the trades as a diesel mechanic.
Love it! A score of years ago, I considered being an auto mechanic after graduating HS but then ended up back in CompSci.
Did you have to go back to school? Did you find a shop that would take you in as an apprentice? And if they did, how did you convince them you can/will be good at the job?
I had lots of personal experience under my belt. I built a hot rod out my garage on the cheap (because I was broke and wanted a nice car). I used that on my resume and they were extremely excited on that. The company I work for is famous for their fleet of "show" semis. It's super super cool and I think the mix of my passion for cars mixed well with their eye for details on their fleet.
However, the bar has never been lower.
I didn't want to do automotive, the piece work is a cancer. You'll do 12 hour days and get paid for 8. Not my cup of tea. I was interested in the big stuff. Offroad equiptment sounded cool too.
I wanted a nice car, so instead of racking up mega debt on a $70,000 mustang I bought an old classic car and learned everything. After 5 years I fully restored it on the cheap (less than $10k) and now I've pivoted my career to that.
A long-ago colleague got a junk classic car and took night classes at the local community college to learn how to fix/restore everything. He finished the degree and quit the tech job.
Anyone currently with a tech job can pay for it out of pocket and barely notice. If it’s something someone thinks they may want to do, they should just do it. Nobody says you have to switch jobs at the end.
And tools have never been cheaper. The knock off chinese clones are used by professionals too. I have tons of automotive friends and they all vary in their level of access to things. I have friends who built their cars on the public road infront of their house, and some friends who took a year long college course.
I'm unconvinced that car mechanic pay is anywhere near programming but either way I'd say you have a pretty select and valuable skill set if you know cars and computers, given how computerised modern cars are.
It depends? It's certainly not gonna be more than you'd make as a SWE/SDE at a big tech company. But for semi diesel specifically you can probably clear $100k+/yr, so you might be making more than some entry level programming jobs that don't pay as well.
It's kinda like law. A mechanic doesn't make more than someone in the right side of the bimodal lawyer pay where 3Ls with the right clerkship or internship walking into a biglaw job paying whatever that is now ($200k?). But a mechanic might make more than the other tranche of lawyers fighting for the rest of the scraps who don't yet (and may never) have a good practice set up.
Fiction has sold AI in the form of Data from Star Trek. A robot with perfect recall of information over a wide range of topics and flawless reasoning.
Today’s AI is nothing like Data with its hallucinations but are taking jobs anyway because it’s “good enough” for many corporations.
P.S. Haven’t been keeping up to date but let’s say I have a story where I retcon a previously an established fact midway through the story with no explanation. If I feed it into AI as part of its training data, will it “challenge” this contradiction? Or will it just blindly accept it? What if the story is part of a prompt, will it “challenge” it in anyway?
I mean even a young child will point out that “that wasn’t what you said earlier”.
I'm looking to leave working in tech. The local public transit bus agency is always hiring mechanics. Going to a technical school for a diesel mechanic training and working there is high up on my list of things to explore. Thanks for the positive reinforcement.
that is so inspiring to hear. would you mind answering a couple of questions:
1) How long were you in software?
2) How did you get your break in the trades? Did you go to school etc?
3) Did you have to start on an apprentice program?
My entire life I've done tech. I remember in middle school all my teachers knew me as the kid who wrote code. I did my degree, and did an internship but nothing ever really stuck (still not sure what happened with that) and my full time job became begging for work.
I didn't do any schooling but thats because I've always liked cars and would tinker at home. So I was very advanced for an entry level. They get government kick backs for hiring apprentices and the less of a burder you are to them, the better. However, the bar has never been lower. Before this, I tried electrician but didn't like it. I have zero experience as an electrician.
Your employer signs you up after you pass probation. Then every year you do a 2 month schooling course which is all government paid.
I love software, but knew I would quickly learn to hate it. I'm not going to be working on my passion projects. I'd be working on horribly boring software used by some corporation.
I really wanted to go into tech because I've been told the trades were the boogie man my entire life.
It's not like the job market was that much better before AI infested every single corner of the market, but it supercharged all of the worst aspects of everything. I've seen people supposedly smarter than I advocate for just giving in, conceding to AI coding as it's the future. But doing so means tossing out my friends who make art or the people who work their asses off to properly test and review code or the writers pouring all of their energy into even mundane dialogue. It means throwing out my dignity as a software engineer, as someone that truly gives a shit about security and code.
Don't let yourself get attached to any tech stack you are not willing to walk out on in 30 seconds flat if you feel the heat around the corner. That's the discipline.
Yes, sadly the most principled will have to contend with the world passing them by when it turns out those principles are dead weight to capital momentum. You could have made the same point during the dawn of the industrial revolution: that purchasing any product of a machine would be betraying your friends who have spent years honing their craft, but either way, people will still opt for the machine because it's 10x cheaper and faster.
AI is good enough now that you can't claim that you aren't using it because you're upholding some higher standard of quality. It is simply a matter of it offending your sensibilities.
That scene is gold. Neil pauses for a little longer than 30 seconds before he decides to abandon Eady, meaning he followed the intent of his rule (walk away), just not the timing (30 seconds flat). If he made the same decision in less than 30 seconds, he probably would've gotten away.
Also, never, ever give more of a shit about things than the people leading the effort. Doesn't matter if that's a for-profit enterprise, a FLOSS project, or just a way to pass the time.
You cannot and will not force them to care more. Don't waste your time on Earth trying to do so.
This may seem all well and good, but let's consider some basic logical mileposts and outcomes:
There will continue to be a glut of available software engineers and techies. Some will, maybe, transition to an AI field; some will get disgusted; some just won't be able to get work.
Jobs of some sort in tech might possibly be available, but wages for the majority of them will go further and further down until they become roughly equivalent to the average minimum wage, if they are not outsourced entirely. Many people will attempt to transition to a non-tech field if the number of available jobs and the wages are not commensurate (especially considering the cost of education). The most desirable of those jobs will also have an upper limit of positions available, and that is of course not paying attention to how many of those will be offloaded onto automation and/or AI. Little things wind up mattering (like the lowering crime rate in California towns suddenly putting auto and window glass repairers out of business) and people who leave tech for other jobs will be fighting for the same dwindling work, with people who are often less difficult to find or work with. Rent won't really go down, and the price of other things will likely continue to rise or stagnate (like many tech salaries; a small percentage of salaries went way up, and the majority went down or are stagnant also). Not saying AI will push everyone out of every field, but it feels like people are thinking in too little of a macro sense.
As AI 'knowledge' is populated by more and more countries with different languages and priorities, English- and some other language speakers will be squeezed out. Probably moreso if and when brain-machine interfaces become de rigeur. Countries with populations of a billion and large families will simply cancel out some people in places like America because social networks will merely favour different people. If my name sounds like yours ethnically, I am possibly far more likely to favour you in a queue. Especially if I am from the same country. This works against people all the time now in the opposite direction. Yes, AI models do use data like this, just as people do.
It is not just tech, of course, and that is the kicker. Tech writing, sure, but also movie and fiction writing, fields dealing large data models, accounting and pharmaceutical research will be largely automated and researched with AI models. Will we need forensic accounting once a model exists?
To the commenters who wrote about how, yes, sure, there will need to be people overseeing things, how do you propose to police that when the lower level AIs skills are so far beyond even the current most senior intermediate or advanced/senior people, and they ramp up so fast, but lack any error correcting? Maybe the AI won't want you involved. Maybe you cannot tell if it chose a good solution or not.
Many... well, no, most good (not even talking godly) tech people only get good by experience, hard work, repetition and the ability to see patterns in their debugging, crashes, program execution, the way their data farms 'feel' (how else do I put this? if you know it feels not right, and sure enough something breaks), and lord knows, even human-computer interaction.
At some point, looking for work is something AIs will discourage us from doing, if they don't already, just for feeling like maybe we won't choose the same solution (would we know?).
Not paranoia. Mere logic. We are attempting to create models but we lack the solutions ourselves. Are we not, like, pricing ourselves out of our own careers (and planet?)?
> I've seen people supposedly smarter than I advocate for just giving in, conceding to AI coding as it's the future. But doing so means tossing out my friends who make art or the people who work their asses off to properly test and review code or the writers pouring all of their energy into even mundane dialogue. It means throwing out my dignity
Conflating things in this inflammatory way is a big mistake. Using a technology employers want you to use is not betraying your friends. Not everything has to be a culture war front.
The author is coming from the games industry. As an arts based industry, AI is EXTREMELY divisive. The author really will lose friends over even so much as touching AI in certain ways, because the artists that built the games industry were already badly abused, and now, they're being squeezed out entirely. For business developers, AI is somewhat less existentially terrifying, as it can be seen to be really empowering to an experienced user.
In the games industry, AI usage immediately eliminates a human job. Why pay a pixel artist if AI can generate 100's of unique little people pixels in seconds, and output them in the right format? Hollywood is going through the same thing: the companies that are building AI for Hollywood have to do so in the bushes, hiding. You don't see them advertising or flashing cash. That's because no one involved in using their wares wants anyone to know they're using them, lest they alienate the highly-talented people they still need to fill the gap between concept AI and full theatrical release AI.
In the software world, we are worried about AI. In the creative industries, they are absolutely pants on fire, screaming at the sky, burning down the village terrified of AI.
> Why pay a pixel artist if AI can generate 100's of unique little people pixels in seconds
Because someone who knows something about pixels needs to make a judgement. It is rare to see a machine-generated artifact (picture, video, text or code) that's good on the first try. And not always a non-specialist can see the issue.
Same thing why we still need human software engineers, even though a machine can generate code. Someone with actual understanding of the problem needs to make a management decision. Just like engineers see code slop (design or implementation) that laypeople vibe-coding don't recognize, artists see the visual slop where layman eye glances over.
Honestly, IMHO, this whole panic is artists' own creation. Instead of educating others on how to spot the issues (and thus reaffirming that expertise still matters - nothing had changed, and probably nothing ever will), a notable fraction went all-in on neo-luddite ideas, as if they don't know the history of their own craft and adjacent creative industries (I guess many really don't, or at least it doesn't click). Evaluate new tools, make use of them when they provide value, skip them where they fall short, and most importantly reaffirm that fancier brushes don't replace the artist in a human society - this is an already well-tested and proven strategy. Ring the existential bells when we'll get to the question of machine cognition rather than just intelligence.
Same for the engineering. Don't shy away from new tools, use them where they're a good fit, don't waste time when they are't (but periodically check out if something changes), explain everyone why you still matter - just to push back on unfortunate misconceptions.
The fact that a lot of companies' upper management went delusional and decided they want to replace humans witch machines (as if don't need responsibility anymore) doesn't help. But - hey - already plenty of stories how it bites them back, so while this period sucks, it's not exactly fucked, just in a state of (a pretty much expected) confusion.
Dishing out pixels or lines of code got somewhat cheaper. Expertise cost remains the same, though.
>Because someone who knows something about pixels needs to make a judgement. It is rare to see a machine-generated artifact (picture, video, text or code) that's good on the first try.
Not if nobody cares for the end product that finely.
And even if it was true, one person can make the judgement, while automation erasing 5 others that would have worked in both the judgement and the graphics wrok.
> Not if nobody cares for the end product that finely.
Yes. But a non-negligible number of people do care. If nobody would've cared, we wouldn't have this drama.
> one person can make the judgement, while automation erasing 5 others that would have worked in both the judgement and the graphics wrok.
Yes. In other words, new tools had increased people's performance for mechanical work - individual units of that can be arguably done faster than before. So hypothetically one person can do a work of five. Note of that erases those people and their skills. The real underlying issue is that demand for it doesn't catch up because world's coincidentally fucked (through a series of unrelated issues, such as a quite few global conflicts) and economies aren't exactly thriving outside of a few niches du jour.
I strongly suspect that if we would've had a flourishing economies around the world, the demand would ramp up and artists (and engineers, and writers, and everyone else whose performance could've been positively affected by new tools) would be in greater position than ever before.
>So hypothetically one person can do a work of five. Note of that erases those people and their skills.
It just doesn't magically take them out of the universe or turns them into unskilled persons.
But it does magically erases those people with those skills as needed employees.
>The real underlying issue is that demand for it doesn't catch up because world's coincidentally fucked (through a series of unrelated issues, such as a quite few global conflicts) and economies aren't exactly thriving outside of a few niches du jour.
Why would that demand have to "catch up"? Just because we can do something faster or automate parts of it doesn't guarantee demand will go up, even in a good economy. No shortage of jobs that vanished forever in a similar even, despite the economy going otherwise up. Even more so now, where it's fucked up anyway, of course.
Everyone in the games industry is badly abused. Mostly because so many want to go into it. And yet game publishers continue to struggle with profitability because budgets have ballooned. I honestly think there's just not enough room in the market for the number of game studios we have today, at least not unless management improves and gets costs under control.
What gets me is that LLM writing has an inhuman voice. Of course we know the tells. Not only this, but that. You're exactly right! —
The woman and man AI voice endemic to YouTube and ilk is also tremendously off putting. M5Stack has a bunch of these videos, and it devalues what they're doing.
And then... Transformer "art". It is some of the worst drek I've seen. I smell it a mile away. It's easily seen by slop english-like characters. Or too glowy humans. Or overall fake feel. For pixel art, I can perhaps see it. But for anything it just feels... Gross.
I'm completely sure management LOVES it cause its cheap and devalues humans.
> I'm completely sure management LOVES it cause its cheap and devalues humans.
And the tragic bit is that instead of educating them about the pitfalls so everyone's on the same page, so a lot of brave hotheads are literally calling for a class war
> And then... Transformer "art". It is some of the worst drek I've seen. I smell it a mile away.
Hating that every third or fourth ad now is AI-generated. So much worse than what entry-level graphic designers can put out while putting them out of work.
Spend some time learning about the concept of the excluded middle[1]. It is something that will be useful on your life going forward, it might even save you a lot of trouble.
I am writing a post about that for my blog but it's still not there so ... check back later? Lol.
Anyway, in this particular scenario, a co-worker not being my friend does not mean we are enemies.
Just to +1 the sentiment here; AI in the gaming industry is absolutely a culture war front. Especially in the indie space, you will lose advertising deals with many sponsors & influencers, for example, if you appear to endorse AI. Probably won't be that way forever, but it's a hotbed right now.
Meanwhile, Asian studios are much more open to experimenting with AI workflows, and will eventually build the next generation of Niantics and AppLovins if this neo-ludditism continues.
Heck, 50% of all Japanese game studios [0] along with all of China's largest studios [1] now use AI within their development pipeline - often with explicit state backing.
You may not like Blizzard or Ubisoft but Tencent, Sony, miHoYo, and even Nintendo are much worse from a work culture, compensation, and work expectation perspective.
> Why should that be a relevant concern? It's not like any of us will be working in game dev if AI wins.
Becuase AI is not taking all jobs. Yes a lot of redundant work will go away, but there is still a real need for human intervention, monkeypatching, and ingenuity.
The North American gaming industry only exists because the entire ecosystem from AdTech to Engines to Marketplaces exists to develop, finance, and distribute IP.
If you stay frozen in the past, you eventually get outcompeted and the ecosystem will leave. And unlike the automotive industry, game devs aren't a core voting bloc.
This is what happened to the entire animation industry and is what is happening to the film and television industry. Gaming will be the next IP driven industry to leave if everyone remains frozen and opposed to innovation.
> Maybe (sic) Chines sell tools for artists pitching them as useful rather then "haha this will make you obsolete you looser look at slop I made" middle finger pitch?
The obsolence and cost saving message is true though and used all over Asia - even in China [0][1].
Either you innovate and compete, or you will get trounced. THIS is the cultural mindset back in Asia.
I dunno, maybe OpenAI and Anthromorphic should learn from Chinese companies how to not act in ridiculously off-putting ways?
Maybe Chines sell tools for artists pitching them as useful rather then "haha this will make you obsolete you looser look at slop I made" middle finger pitch?
This is seemingly spoken from an ignorant and insulated position. The victims of invasion don't get to decide whether or not they live on a war front, nor do the countless skilled and creative individuals losing their entire careers almost overnight.
By the way it's class war you're talking about, not culture war.
Propertly testing and reviewing code is a must with AI, it's actually more important. I don't know why the author feels like it's assumed that this goes out the window with AI
Not sure the technology is the issue here. He isn't debating using Java v C#, and one is a betrayal of all that is holy, so he refuses to use it.
I think everyone kind of feels that AI is sucking up all the content that we have all created collectively, and we all know that the bell is tolling for thee, no matter how much you adapt. So if you see friends being fed into a meat grinder, you can have a 'culture war' take on it. He's posting on line to vent, something everyone is venting about.
Of course, for a job, to get a paycheck, we'd use any technology, even if we are the ones running the meat grinder.
I think you might not get the context this is coming from, it’s about a game developer laid off from Blizzard not wanting to partake in the technology that will justify layoffs of their friends. Pretty straight forward to me.
You will not have a job in software in the future if you refuse to use AI.
I fully stand behind that prediction.
edit: despite the downvotes, I'll double down: most of today's software jobs will disappear. Your job, if it is in software, will disappear. It might transform into something new, if you're lucky. Or it might just go away entirely.
I guess I'm just confused by this sentiment. Are you making these conclusions while considering the fact that AI is still heavily subsidized? The economics of AI isn't quite the same as other software/tech.
I don't think it's going anywhere, but I don't know what happens when prices start to rise because these companies need to start turning a profit.
I'm happy using GLM 5.2 at API pricing. There will never be a point where essentially unlimited usage (i.e. what I, as a normal developer, can use at peak effort) of extremely good models will become financially unattainable.
> There will never be a point where essentially unlimited usage (i.e. what I, as a normal developer, can use at peak effort) of extremely good models will become financially unattainable.
The way you phrase this prevents refutation. But there will be a point where ordinary individuals cannot participate in the majority of the upside.
In the future, some class of models will require enormous compute that is outside the financial capital capability of ordinary individual contributors, middle class, and upper middle class. This will be sold as a capability to well-funded companies.
> For the vast majority of people who currently have a job in software... you will not have a job in software in the future whether you use AI or not.
The tide is coming for almost all of us.
I'll agree with you as soon as all video games are in the metaverse and run natively on the blockchain.
AI does your job. Robots do your job. These are real and substantial and actually provide enormous value. You can get more done per unit of time.
I want them to do my job and everyone else's jobs, I just want to make sure there's a functioning economy we can participate in and benefit from at the other side.
Corporations are the organizational equivalent of AI, they one thing well asmass money, their output aside from that is total shit. Not only are you betraying your friends/community/loved ones but such a setup is necessary to prevent the total collapse of society.
yeah, the question is what would you do with $100 million? I mean I don't have $100 million but I'd give it to the arts after I've taken care of myself well and science. Am I gonna make $100 million playing with AI? I don't even know if I can take care of myself, shit, but the government ain't gonna do that. What are the new jobs after the Saaspocalypse? Well all the money went to those companies and what do those companies want or what do those newly minted millionaires want? Hopefully something I have but shit yo. So they're gonna go back to where they came from and be millionaires but how much is Dave Chappelle changed the city that he moved to? Is there a Dave Chappelle center for comedy? some of them want ASI so I look forward to studies on human intelligence. Maybe they'll pay human subjects to model supply two digit numbers while in an MRI machine hopefully enough of those studies to pay my mortgage.
> I mean I don't have $100 million but I'd give it to the arts
I might be naive, but rather than giving money "to the arts", I would much prefer to give money "to the people" en masse, and then leave it up to everyone to decide whether they want to make art for art's sake.
A month ago, I fell back into reading patio11's "don't call yourself a programmer" and I found it fitting. The core of the message wasn't about the title we assign to ourselves but the "other career advice".
I felt compelled to write "don't call yourself a Software engineer" [0], because we are still falling into the same trap of thinking we are hired only for our technical skills.
If we are just looking at a skills and these are assessed by parsing through a resume, then OP is right. We are all at a disadvantage. But the job search starts way before you are looking for a job. It's all about the connections you make along the way.
I feel like this advice is not very useful because when you call yourself a software engineer or programmer, you are doing it in order to sell a service.
Your customers are companies looking for someone to slot into a box called "software engineer" and so you sell yourself as such. Nothing wrong with that.
We should also note who Patrick was at the time. He was an SEO consultant and in general a business development expert. It just also happened that he was able to code. And he was very very early to the field. An SEO expert was barely a thing.
So if your only skill is software development, then of course you would call yourself that. And if your main skill is SEO or some other marketing channel, then you call yourself that.
I think the real takeaway from the advice "don't call yourself a programmer" is to search the market for higher paid opportunities, where you can still leverage coding. And you can call yourself a programmer while doing so.
>Engineers in particular are usually very highly paid Cost Centers, which sets MBA’s optimization antennae to twitching. This is what brings us wonderful ideas like outsourcing, which is “Let’s replace really expensive Cost Centers who do some magic which we kinda need but don’t really care about with less expensive Cost Centers in a lower wage country”. (Quick sidenote: You can absolutely ignore outsourcing as a career threat if you read the rest of this guide.) Nobody ever outsources Profit Centers. Attempting to do so would be the setup for MBA humor. It’s like suggesting replacing your source control system with a bunch of copies maintained on floppy disks.
Two weeks ago, I got 100/100 of a test from a big company for a first screening without using AI. I was pretty confident that I would pass the first round, even hinted few of my friends, but ended up being rejected with an automated mail… The job market is insane at this point and I am not sure what the recruiters are actually looking for. If the candidate uses AI they’re eliminated, if not they’re eliminated. I guess this is one of these times we read on history books: great unemployement.
Nowadays more than half of job postings are fake. It's either process to show activity or they already have one to hire but need to follow established process.
Getting through an interview process during a bull market has always had a small random component. I think we all have to understand and accept that in bear hiring markets, almost the entire component is random. Having a perfect skill quiz or hackerrank score and getting rejected should not cause you to try to figure out what you did wrong.
> The job market is insane at this point and I am not sure what the recruiters are actually looking for.
To justify their own jobs.
In theory there's never been a better time to hire on SWE talent. There are lots and lots of candidates who rode high during the COVID hiring wave, took on debt based on a high income, got fired, and now need money.
But hiring isn't picking up. You have a bunch of people in the HR industry who realize that for the most part, the combination of candidate filtering, ML, and a basic tech interview process could probably do their jobs. So they have to make the process as byzantine and difficult as possible to be able to go to the suits and say "look at all of these low-quality candidates we kept out!"
> I guess this is one of these times we read on history books: great unemployement.
The good/bad news is that if this continues, there will be either a regression to the mean or a massive de-stablization of most societies. You can't kick most of the working-age population out of their jobs.
I was just involved in the hiring process for a new rec on a team adjacent to mine, and the HR/screening process was a complete mystery as to candidate filtering for outside applicants. So many AI tools making arbitrary judgments from resume content to psychological profiles of candidates, led us to 0 organic people “qualified” enough to interview for an SRE/Devops role. All we hire now is referrals as they are the only ones to get pass automated screening
The industry will realize that while getting LLMs to write code is easy, getting LLMs to write good, production ready code is a skill all on its own, which simply must be done by a human and is not automatable to an LLM in any sense effectively. That will be the differentiating factor software engineering in the future, I think.
If I'm being blunt, if you are in the game industry, you probably have nothing to worry about in terms of LLM coding replacing you, because the tooling used in the gaming industry is as unfriendly to LLM coding as it gets: Heavily visual scripting based, extremely reflection heavy, and the code, Unreal C++ and Unity C#, looks like regular C++/C#, but doesn't behave like normal C++/C#. LLMs simply cannot reason about hidden implicit states effectively, so if the code looks right but doesn't act right, LLMs will simply get confused and start hallucinating.
I've spoken about my troubles getting as much mileage out of LLMs as before while being vague about my industry and, well, this hits the nail in the head.
> because the tooling used in the gaming industry is as unfriendly to LLM coding as it gets: Heavily visual scripting based
At the start of the LLM craze, we (as a team) put ChatGPT to test with Godot. It wasn't very successful in that, IIRC, GDScript 2 was just released and ChatGPT's training corpus was so obviously based on GDScript 1.
We could make ChatGPT outline steps of how to accomplish things in Godot, sort of like getting a customized tutorial. When code isn't involved, ChatGPT was okay and Gemini seemed to fare better. Based on vibes, I think Gemini had a marginally better handle on GDScript 2 for some reason.
I've been trying it lately with Claude still with mixed results. I had to install a few skills/extensions for it (can't tell you which as I basically just blindly followed what our AI advocate recommended). Sometimes it works but when it doesn't it's harder to put a finger on why. Overall I prefer the DX of generating customized tutorials with ChatGPT.
> extremely reflection heavy
Big time. And IME we don't even have to deal with textbook reflection here. Game entities are so convoluted (remember the Fallout 3 Train NPC, the stories of how Skyrim works, etc.) that it is really pushing inheritance and OO to contortions it shouldn't be doing.
Dirty confession: in our game we have this GIANT switch-statement dealing with game objects. It happens in a handful of places, for different game object types. LLMs (Copilot and Codex) could generate the monkey code of adding a switch case and even writing the body but sooner than later, when the new objects have to interact with others, LLMs just can't reason around it. Not to mention the hundred edge cases you have to consider!
And before some smart-ass comments: in my almost-decade of dealing with this code base there have been a handful of attempts to "refactor" these switch-statements, always a newcomer's enthusiastic effort. I'm proud to say, I've managed to slay one of the giants, the only successful effort to my knowledge, and this only happened last year. But I did so by basically delegating to another of the giants; they turned out to be twins and we could do without one. The dirty way is the clean way because the alternative contains Lovecraftian geometries.
Claude has been great for finding edge-case bugs but that's only once the code has been written properly. Generally if QA reports a bug on a pre-release feature, it's at most 50/50 if Claude can debug it. But if it's a player report/live incident, I'd say Claude's chances goes up to around 80%.
All that said,
> if you are in the game industry, you probably have nothing to worry about in terms of LLM coding replacing you
Hah-hah. Look, I agree with you but please get in touch with upper management. As an engineer I'm confident on the value I bring to the table but I'm not sure management "gets" that. Like, no matter how I tell them what features I shipped, what infra I built, etc., it doesn't come across to them that LLMs would not have been able to automate that output!
I am currently quitting a company of 10 years of employment. And I keep hearing how everything's shit. Btw. I'm located in Tokyo where it isn't as bad, apparently, but...
Let's see. My plan:
- Have my own company and start looking for customers. (Rust consulting)
- Keep looking for job opportunities, but don't succumb for shit jobs.
It might be that I'm too hopeful, but you can't know unless you try.
Anyway, I may join the "everything is shit" crowd in half a year if nothing pans out, but until then, I'm hopeful.
That might be a truism that's generally applicable, but some people have enough of a safety net to last them several years while building a business. We don't know OP's circumstances.
There are a number of reasons I’d site for the current job market tightness:
- political: there’s an enormous amount of uncertainty here. All businesses make plans and uncertainty puts them all on pause.
- economic: related to political, but we’re teetering on a very bad recession. Watch where national oil reserves go.
- AI: I throw this in with every new technology that comes out. There is always a period of chaos before normalization. We’re still in the chaos phase.
- Business Pain: Right now I don’t see any sector that’s in pain. Inflation has hurt consumers, but we’re still spending. When consumers lock it down, that pain comes back and job market shifts with it.
I have no solution other than figuring out a way to do your own thing. There’s no better time to be a founder.
- GenAI becomes a foundational requirement for tech and non tech sectors. If you’ve refused to engage, you’ve self-selected out of any of those sectors.
- GenAI usage shifts down to just the tech sector, but in an integrated fashion where current engineering practices are still desired. Everyone survives, but pay scales are adjusted down by a not-insignificant amount.
- GenAI bubbles badly, OpenAI and Anthropic merge with Google/Microsoft/Oracle/IBM/???. Tokens become extremely expensive and no one is leaning into agentic integration. Everyone thrives.
It might also be foolish, but I'm mid-pivot to becoming an Actuary. They credential through a very transparent exam system, and interviews are relatively cursory, assuming you've passed the appropriate exams on the right timeline. I've got a math degree, and my software experience is all in the data engineering space, which seems to be in demand.
I got laid off from a large company last summer, and took some time off to travel. Now I've got a chill, low paid dev job in a resort town in New Zealand, but my sense is that dev work is not going to be the thing for me long term. This job will pay my bills while I pivot, but they're not going to sponsor me to extend my visa, so I'm on a bit of a timer. The market back in the US seems like an ongoing mess and I don't want to get back into it.
I work in an industry tangentially involved with the ML build-out (think companies like Broadcom, Marvell, etc.). We can't find enough people, if you have like 3+ years of experience with PCIe, Ethernet, DDR, you're a shoe in. Verification, Validation, Design, Customer Applications, Firmware, you name it, we need it. The pay is good too, especially for people who got in a year ago or more, stock base compensation has taken off like a rocket.
Hiring here is a little bit more old school, I guess? Especially because the types of roles we are hiring now are usually 5+ years of experience, we focus more on learning about what the candidates have done in the past, the leetcode type of question interview is just a small part, and matters more for prospective Jr. hires.
That being said, we aren't hiring that many fresh graduates anymore, we already have some we hired, we're focused on investing in them, getting them to learn more about our hardware and code, etc. and hoping to retain them.
I feel like any advice you get from someone would be as useful as "how to be a good coin flipper" advice from the 1 person in 1024 who flipped heads 10 times in a row. In other words it would be purely survivorship bias.
I joined a new company 6 months ago. I interviewed at 16 companies and got 5 offers from a mix of ai cos / big tech / trading firms
Background is SWE at an AI co that's in the news sometimes
It felt about the same in terms of grind effort from my last search in 2022. the main difference was ai companies cared a lot about your understanding of agentic systems and harness / context engineering, and had much more practical rounds with less leetcode (usually 1 medium). More legacy firms (finance / some big tech) still expected you to solve 3-4 leetcode medium/hards throughout the process
Anecdotally I've seen the latter to be true and have had third-party recruiters echo that familiarity with AI use in coding has become a pivotal part of job interviews with startups.
That being said, I'm not sure how much job security having such prowess would convey because I feel AI will be better than us at that too eventually (if not already).
I'm starting a new job in a few weeks, and can confirm (for startups at least) experience with coding agents is something companies are looking for. Multiple companies I interviewed with had a AI assisted interview session to go along with a more typical closed book programming session. I was asked about my use of coding agents in behavioral interviews. I'm not an ML guy, just a generalist SWE with 4 YOE. I only got one offer in my search, but it only took ~a month and I feel pretty good about being able to get more offers with more searching. It helps that I'm young, no dependents, and willing to relocate.
My perspective is it is impossible to cut through the three layers of bullshit between you and anyone who knows what they are talking about. The only way to do this is with brand-name qualifications, like "MIT graduate", not things that are actually impressive. This is also why you see senior developers saying, "the offers I'm getting are bigger and bigger," meanwhile skilled younger developers need to become a marketing professional just to get an interview.
Recruiters have utterly given up on being efficient in the market. I do not know why, but there is something very wrong given "spamming the same brand-name fish all the other recruiters are spamming" is their only strategy. My guess is there is a combination of bad (or an entire lack of) hygienic data filtering and a disconnect between compensation and terminal goals (hiring the best candidates).
What do you mean by easy? Do you mean FAANG or equivalent salary? What level of seniority?
Can speak to my experience that if you are a senior engineer in London the market is relatively easy at the moment (or was at the beginning of the year) even with no connections.
Silicon design/verification. In really high demand at the moment, I guess because of the death of Moore's Law - now it is much more worthwhile making custom chips.
Easy hires are the same now as in the past. They have held several specialized roles before without letting it narrow their career path, and they have at least a decade of experience.
The problem with specialized roles is that nothing lasts too long in software. Given enough time in it, nobody really has an edge. Everyone is smart enough to have invented and implemented the very thing eating the world right now. They just don't have supervillain money or clout, so they work for you instead.
OP's opinion about AI coding is pretty obvious in this blog post. Maybe some of that sips out during interviews which certainly will spook the employer.
In the game industry this is actually a positive thing. If you think that AI can write code better than you can, I'm not going to disagree with you, but maybe you should try getting better or more specialized at what you do, because I can ask the AI simple questions about my field and it's completely wrong 3-4 prompts in.
Using this technology to build anything remotely serious, OR EVEN GAMES, is wildly stupid.
I'm not used to seeing informed takes on the games industry on Hacker news, but then I looked up to see your name. Love your stuff, your website was mind opening when I was in college.
If you don't have a full time job for a year, why don't you start a software business on your own? It's hard to succeed, but, successful or now, you will have something to show people and something that you are passionate to talk about. It will give you a new way to connect to people, and quite likely will help you to get hired. And who knows, maybe you will succeed.
100% this. Interviewing isn't something that can compound. Striking out from company after company doesn't leave behind a trail of real work and real lessons. Starting a business is tough but it really does teach skills that are hard to find any other way (about sales, recruiting, management, etc). After a certain point, it's wiser to give up on getting hired, and just hire yourself and build something.
I wasn't there but this seems like the same feelings people would have had in the Rust Belt when the first factories started closing and getting a job started getting harder.
During the pandemic money printing things got very weird. It created a lot of leverage and bullshit companies and bullshit dev work which led to artificial demand for software developers.
We are still in the post-pandemic hangover.
If you look up M2 money supply on St Louis fed - that chart has more influence on the job market in the US than anything.
The macro whiplash compounds this problem for people like OP in a few ways:
- cheap money leads to hiring frenzy (cheap capital costs lead to investments in human capital in software)
- developers get conditioned to artificially high demand and assume it will be like that forever
- artificially high demand attract people into software dev for the money instead of love of the art (increasing supply)
- when capital gets expensive again companies have to correct for over-hiring with layoffs and hiring freezes
- developers are stuck in a market with crashing demand (because of higher cost of capital) and over-supply (people attracted to work when cost of capital was cheap)
Everyone says it's about AI, but AI is more like the flavor & scapegoat, the substance is all a consequence of macro policy.
The next time the fed does quant easing labor market will kick up again.
Nobody likes to hear it, but this is the only explanation that makes sense. We had an unprecedented economic shock, and we're dealing with an unprecedented economic fallout. The only question is how much longer it will last.
ChatGPT wasn’t released until the end of 2022, and it wasn’t until the next spring where it really started to take off. Spring 2023 is when ChatGPT started to gain traction as a direct to consumer app and it wasn’t until later in the year when startups started building on top of it.
Just use python for all practical programming problems. Lists, sets, and dicts are all you need for most leetcode problems; dynamic typing is convenient; there's good ergonomics for http and other random utility tasks; and pretty much every company is cool with python in an interview. You'll probably only see language trivia questions for languages you claim a specialty in (there's a huge market for C++ specialists, for instance).
The blog lacks crucial info. What type of projects have you worked on? What work are you applying for? Personally, "small contractor" and Blizzard does not translate well to the typical "enterprise web dev" role.
The first place to look for jobs should be in your network, people that worked with you, teachers, ex-managers.
Applying for jobs out of the blue usually sucks. In the ideal world, you want recruiters calling out to you.
Don't assume you can't do proper software engineering using AI. You can. The people that want to create loops are not the only ones delivering with acceptable productivity. Lot's of us still write code, at least interfaces, traits, modules or whatever, and just use the AI to fill the blanks on the really tedious code.
The job market to be honest has been very fucked. To me a lot of this sounds like people experiencing how terrible tech hiring has become for the first time after being in a stable job for a long time. Almost everything the Author said, was something I’ve experienced when I was laid off in the 2022-2023 wave of layoffs. At the time I was told “it was a skill issue”.
It really is awful right now. I'm lucky enough to still have a job, but floated my resume around earlier this year. I have a pretty good resume and and 15 YOE, and got turned down EVERYWHERE. I used to at least get interviews at like 50% of places I applied to.
And then recruiter spam is COMPLETELY gone. I'm not really complaining, but it feels indicative of where things are at.
You're right on the latter point. I always wondered if Linkedin increased their spam filter or make it 10x more expensive to DM candidates, but I guess it's simply because there is simply a lack of demand right now
I'm transitioning to freelance (independent consultant, full stack and Cloud), and I find I'm working alongside AI in business manager's hands. It's an uncanny mode of competition.
I agree that the online hackerrank quizes where it isn't even a video call is dumb because so many people cheat and if you don't, you're at a disadvantage.
Lots I agree with here, but...
> I would promptly remove them from my LinkedIn connections afterwards because I'm exhausted of pointless connections and recruiters.
Why would you do something like this, it's just counter productive. I've had numerous recruiters reach out weeks or months later to say "hey another team is interested", or even when they have moved on to other jobs.
Stop being so bitter you're just shooting yourself in the foot.
I'd suggest not adding recruiters as LinkedIn connections in the first place. I mean, LinkedIn is hardly sacred, but why are you adding people you don't know as a connection?
Recruiters only add you as a connection if they can't afford LinkedIn premium, which is what you need to message people you're not connected to (except for connection requests). That probably means that they're not very successful recruiters.
I heard that recruiters also use amount of connections to filter out candidates, and they probably view it as if you don't have 500-1000 connection probably you just don't fit basic requirements of the the role even with 10+ yoe.
Even is so stupid but looks like in last few years lot of strange metrics like that used more and more.
If you truly give a shit you have to change and help make the mess less mess. It sucks, it might be worse than it was, but you can't continue giving a shit by not participating. The horse has left the barn on this one.
The frontier model companies could all collapse tomorrow but the tech is not going anywhere.
This is peripheral about bearblog, but it's so grating to see the "D M, Y" date format with the comma. The correct format is "D M Y." It's like someone deciding to write June, 6, 2026 for some reason.
After 5+ years of actively trying to get into the field (pre AI), I left.
I threw my degree in the toilet, I closed my linkedin, and I went to go work in the trades as a diesel mechanic.
Greatest choice I've ever made. The pay is great, the work is steady, the coworkers are relaxed and not trying to one up each other. I'm now being paid to go to school, and get raises every year until I'm fully ticketed (way more than I ever made in the entry tech positions).
I've heard non-stop my whole life that if I join a trade it's going to be grunt labor and I'll be paralyzed in 5 years. Maybe some are, but this isn't hard at all. I lay on a creeper and turn wrenches. Anything over 50lbs we have lifts for.
Tech has become fun again, I'm just making projects because its what I wanted. I come home and relax by writing on my projects.
Now, I'm watching my tech friends from a distance and my only regret is not doing this sooner.
"Sometimes you gotta give in to win"
Reading this and some of your other comments in this thread, I think it’s awesome you’ve landed up doing what you really enjoy and are well compensated for it.
It makes me wonder if I would be happier doing something else, but (because of my personality) I’m very doubtful.
Since you see yourself as also being a computer guy I’m assuming that lack skill or intuition was not why you left the industry, so don’t read the rest of my comment as talking about you.
But I’ve definitely seen plenty of people in the software development industry where they may get by “okay” at their job, but things don’t tend to “click” as easily (in terms of intuitive understanding) for them the same way they do for me.
So I feel lucky and deeply happy to be at a company I enjoy working at and doing what has always been my passion.
It’s not that the computer industry is completely terrible (although plenty of parts of it certainly are), it’s just that for some people it’s not their true passion (which is fine).
You might be able to be a diesel mechanic and clean up in tech. This video is almost a year old, about applying Claude Code to boring businesses, a mobile diesel mechanic service in this case study: https://www.youtube.com/watch?v=gWNFna6fgS8
> I closed my linkedin, and I went to go work in the trades as a diesel mechanic.
Love it! A score of years ago, I considered being an auto mechanic after graduating HS but then ended up back in CompSci.
Did you have to go back to school? Did you find a shop that would take you in as an apprentice? And if they did, how did you convince them you can/will be good at the job?
I had lots of personal experience under my belt. I built a hot rod out my garage on the cheap (because I was broke and wanted a nice car). I used that on my resume and they were extremely excited on that. The company I work for is famous for their fleet of "show" semis. It's super super cool and I think the mix of my passion for cars mixed well with their eye for details on their fleet.
However, the bar has never been lower.
I didn't want to do automotive, the piece work is a cancer. You'll do 12 hour days and get paid for 8. Not my cup of tea. I was interested in the big stuff. Offroad equiptment sounded cool too.
Did you have qualifications / an interest in this field before switching? What drawn you to this occupation in the first place? Just being curious
I'm a car guy, and a computer guy.
I wanted a nice car, so instead of racking up mega debt on a $70,000 mustang I bought an old classic car and learned everything. After 5 years I fully restored it on the cheap (less than $10k) and now I've pivoted my career to that.
A long-ago colleague got a junk classic car and took night classes at the local community college to learn how to fix/restore everything. He finished the degree and quit the tech job.
Anyone currently with a tech job can pay for it out of pocket and barely notice. If it’s something someone thinks they may want to do, they should just do it. Nobody says you have to switch jobs at the end.
100%
And tools have never been cheaper. The knock off chinese clones are used by professionals too. I have tons of automotive friends and they all vary in their level of access to things. I have friends who built their cars on the public road infront of their house, and some friends who took a year long college course.
We all ended up in the same place.
I'm unconvinced that car mechanic pay is anywhere near programming but either way I'd say you have a pretty select and valuable skill set if you know cars and computers, given how computerised modern cars are.
It depends? It's certainly not gonna be more than you'd make as a SWE/SDE at a big tech company. But for semi diesel specifically you can probably clear $100k+/yr, so you might be making more than some entry level programming jobs that don't pay as well.
It's kinda like law. A mechanic doesn't make more than someone in the right side of the bimodal lawyer pay where 3Ls with the right clerkship or internship walking into a biglaw job paying whatever that is now ($200k?). But a mechanic might make more than the other tranche of lawyers fighting for the rest of the scraps who don't yet (and may never) have a good practice set up.
Until AI comes for your mechanics job too,
https://www.tomshardware.com/tech-industry/artificial-intell...
I feel we are getting the worse of “both worlds”.
Fiction has sold AI in the form of Data from Star Trek. A robot with perfect recall of information over a wide range of topics and flawless reasoning.
Today’s AI is nothing like Data with its hallucinations but are taking jobs anyway because it’s “good enough” for many corporations.
P.S. Haven’t been keeping up to date but let’s say I have a story where I retcon a previously an established fact midway through the story with no explanation. If I feed it into AI as part of its training data, will it “challenge” this contradiction? Or will it just blindly accept it? What if the story is part of a prompt, will it “challenge” it in anyway?
I mean even a young child will point out that “that wasn’t what you said earlier”.
I'm looking to leave working in tech. The local public transit bus agency is always hiring mechanics. Going to a technical school for a diesel mechanic training and working there is high up on my list of things to explore. Thanks for the positive reinforcement.
that is so inspiring to hear. would you mind answering a couple of questions:
1) How long were you in software? 2) How did you get your break in the trades? Did you go to school etc? 3) Did you have to start on an apprentice program?
Thank you very much
My entire life I've done tech. I remember in middle school all my teachers knew me as the kid who wrote code. I did my degree, and did an internship but nothing ever really stuck (still not sure what happened with that) and my full time job became begging for work.
I didn't do any schooling but thats because I've always liked cars and would tinker at home. So I was very advanced for an entry level. They get government kick backs for hiring apprentices and the less of a burder you are to them, the better. However, the bar has never been lower. Before this, I tried electrician but didn't like it. I have zero experience as an electrician.
Your employer signs you up after you pass probation. Then every year you do a 2 month schooling course which is all government paid.
Did you go into tech because you love software or for the money?
I love software, but knew I would quickly learn to hate it. I'm not going to be working on my passion projects. I'd be working on horribly boring software used by some corporation.
I really wanted to go into tech because I've been told the trades were the boogie man my entire life.
It's not like the job market was that much better before AI infested every single corner of the market, but it supercharged all of the worst aspects of everything. I've seen people supposedly smarter than I advocate for just giving in, conceding to AI coding as it's the future. But doing so means tossing out my friends who make art or the people who work their asses off to properly test and review code or the writers pouring all of their energy into even mundane dialogue. It means throwing out my dignity as a software engineer, as someone that truly gives a shit about security and code.
Don't let yourself get attached to any tech stack you are not willing to walk out on in 30 seconds flat if you feel the heat around the corner. That's the discipline.
Yes, sadly the most principled will have to contend with the world passing them by when it turns out those principles are dead weight to capital momentum. You could have made the same point during the dawn of the industrial revolution: that purchasing any product of a machine would be betraying your friends who have spent years honing their craft, but either way, people will still opt for the machine because it's 10x cheaper and faster.
AI is good enough now that you can't claim that you aren't using it because you're upholding some higher standard of quality. It is simply a matter of it offending your sensibilities.
Both the people that claim they won't use LLMs and the people that claim they don't even need to look at the code anymore are deluding themselves.
That quote from Heat (1995) is so deployable in so many different contexts.
the irony is that McCauley (DeNiro) doesn't actually follow his macho adage as the movie shows in the end..same goes for real developers.
Thats the point of the movie: the organized and disciplined leader didnt follow his own advice while the hot-head reckless gambling degenerate did.
Chris also goes back, Charlene saves him.
That scene is gold. Neil pauses for a little longer than 30 seconds before he decides to abandon Eady, meaning he followed the intent of his rule (walk away), just not the timing (30 seconds flat). If he made the same decision in less than 30 seconds, he probably would've gotten away.
BTW, 'Heat 2' is pretty good, gives a lot of backstory as well as current-story, mostly about Chris: https://www.amazon.com/Heat-2-Novel-Michael-Mann/dp/00626533...
Also, never, ever give more of a shit about things than the people leading the effort. Doesn't matter if that's a for-profit enterprise, a FLOSS project, or just a way to pass the time.
You cannot and will not force them to care more. Don't waste your time on Earth trying to do so.
>Now, if you're on me and you gotta move when I move, how do you expect to keep a...a tech stack?
> What are you a monk?
This may seem all well and good, but let's consider some basic logical mileposts and outcomes:
There will continue to be a glut of available software engineers and techies. Some will, maybe, transition to an AI field; some will get disgusted; some just won't be able to get work.
Jobs of some sort in tech might possibly be available, but wages for the majority of them will go further and further down until they become roughly equivalent to the average minimum wage, if they are not outsourced entirely. Many people will attempt to transition to a non-tech field if the number of available jobs and the wages are not commensurate (especially considering the cost of education). The most desirable of those jobs will also have an upper limit of positions available, and that is of course not paying attention to how many of those will be offloaded onto automation and/or AI. Little things wind up mattering (like the lowering crime rate in California towns suddenly putting auto and window glass repairers out of business) and people who leave tech for other jobs will be fighting for the same dwindling work, with people who are often less difficult to find or work with. Rent won't really go down, and the price of other things will likely continue to rise or stagnate (like many tech salaries; a small percentage of salaries went way up, and the majority went down or are stagnant also). Not saying AI will push everyone out of every field, but it feels like people are thinking in too little of a macro sense.
As AI 'knowledge' is populated by more and more countries with different languages and priorities, English- and some other language speakers will be squeezed out. Probably moreso if and when brain-machine interfaces become de rigeur. Countries with populations of a billion and large families will simply cancel out some people in places like America because social networks will merely favour different people. If my name sounds like yours ethnically, I am possibly far more likely to favour you in a queue. Especially if I am from the same country. This works against people all the time now in the opposite direction. Yes, AI models do use data like this, just as people do.
It is not just tech, of course, and that is the kicker. Tech writing, sure, but also movie and fiction writing, fields dealing large data models, accounting and pharmaceutical research will be largely automated and researched with AI models. Will we need forensic accounting once a model exists?
To the commenters who wrote about how, yes, sure, there will need to be people overseeing things, how do you propose to police that when the lower level AIs skills are so far beyond even the current most senior intermediate or advanced/senior people, and they ramp up so fast, but lack any error correcting? Maybe the AI won't want you involved. Maybe you cannot tell if it chose a good solution or not.
Many... well, no, most good (not even talking godly) tech people only get good by experience, hard work, repetition and the ability to see patterns in their debugging, crashes, program execution, the way their data farms 'feel' (how else do I put this? if you know it feels not right, and sure enough something breaks), and lord knows, even human-computer interaction.
At some point, looking for work is something AIs will discourage us from doing, if they don't already, just for feeling like maybe we won't choose the same solution (would we know?).
Not paranoia. Mere logic. We are attempting to create models but we lack the solutions ourselves. Are we not, like, pricing ourselves out of our own careers (and planet?)?
It is more than hubris.
The author is coming from the games industry. As an arts based industry, AI is EXTREMELY divisive. The author really will lose friends over even so much as touching AI in certain ways, because the artists that built the games industry were already badly abused, and now, they're being squeezed out entirely. For business developers, AI is somewhat less existentially terrifying, as it can be seen to be really empowering to an experienced user.
In the games industry, AI usage immediately eliminates a human job. Why pay a pixel artist if AI can generate 100's of unique little people pixels in seconds, and output them in the right format? Hollywood is going through the same thing: the companies that are building AI for Hollywood have to do so in the bushes, hiding. You don't see them advertising or flashing cash. That's because no one involved in using their wares wants anyone to know they're using them, lest they alienate the highly-talented people they still need to fill the gap between concept AI and full theatrical release AI.
In the software world, we are worried about AI. In the creative industries, they are absolutely pants on fire, screaming at the sky, burning down the village terrified of AI.
> Why pay a pixel artist if AI can generate 100's of unique little people pixels in seconds
Because someone who knows something about pixels needs to make a judgement. It is rare to see a machine-generated artifact (picture, video, text or code) that's good on the first try. And not always a non-specialist can see the issue.
Same thing why we still need human software engineers, even though a machine can generate code. Someone with actual understanding of the problem needs to make a management decision. Just like engineers see code slop (design or implementation) that laypeople vibe-coding don't recognize, artists see the visual slop where layman eye glances over.
Honestly, IMHO, this whole panic is artists' own creation. Instead of educating others on how to spot the issues (and thus reaffirming that expertise still matters - nothing had changed, and probably nothing ever will), a notable fraction went all-in on neo-luddite ideas, as if they don't know the history of their own craft and adjacent creative industries (I guess many really don't, or at least it doesn't click). Evaluate new tools, make use of them when they provide value, skip them where they fall short, and most importantly reaffirm that fancier brushes don't replace the artist in a human society - this is an already well-tested and proven strategy. Ring the existential bells when we'll get to the question of machine cognition rather than just intelligence.
Same for the engineering. Don't shy away from new tools, use them where they're a good fit, don't waste time when they are't (but periodically check out if something changes), explain everyone why you still matter - just to push back on unfortunate misconceptions.
The fact that a lot of companies' upper management went delusional and decided they want to replace humans witch machines (as if don't need responsibility anymore) doesn't help. But - hey - already plenty of stories how it bites them back, so while this period sucks, it's not exactly fucked, just in a state of (a pretty much expected) confusion.
Dishing out pixels or lines of code got somewhat cheaper. Expertise cost remains the same, though.
>Because someone who knows something about pixels needs to make a judgement. It is rare to see a machine-generated artifact (picture, video, text or code) that's good on the first try.
Not if nobody cares for the end product that finely.
And even if it was true, one person can make the judgement, while automation erasing 5 others that would have worked in both the judgement and the graphics wrok.
> Not if nobody cares for the end product that finely.
Yes. But a non-negligible number of people do care. If nobody would've cared, we wouldn't have this drama.
> one person can make the judgement, while automation erasing 5 others that would have worked in both the judgement and the graphics wrok.
Yes. In other words, new tools had increased people's performance for mechanical work - individual units of that can be arguably done faster than before. So hypothetically one person can do a work of five. Note of that erases those people and their skills. The real underlying issue is that demand for it doesn't catch up because world's coincidentally fucked (through a series of unrelated issues, such as a quite few global conflicts) and economies aren't exactly thriving outside of a few niches du jour.
I strongly suspect that if we would've had a flourishing economies around the world, the demand would ramp up and artists (and engineers, and writers, and everyone else whose performance could've been positively affected by new tools) would be in greater position than ever before.
>So hypothetically one person can do a work of five. Note of that erases those people and their skills.
It just doesn't magically take them out of the universe or turns them into unskilled persons.
But it does magically erases those people with those skills as needed employees.
>The real underlying issue is that demand for it doesn't catch up because world's coincidentally fucked (through a series of unrelated issues, such as a quite few global conflicts) and economies aren't exactly thriving outside of a few niches du jour.
Why would that demand have to "catch up"? Just because we can do something faster or automate parts of it doesn't guarantee demand will go up, even in a good economy. No shortage of jobs that vanished forever in a similar even, despite the economy going otherwise up. Even more so now, where it's fucked up anyway, of course.
Everyone in the games industry is badly abused. Mostly because so many want to go into it. And yet game publishers continue to struggle with profitability because budgets have ballooned. I honestly think there's just not enough room in the market for the number of game studios we have today, at least not unless management improves and gets costs under control.
[delayed]
The beatings will continue until management improves and gets costs under control.
What gets me is that LLM writing has an inhuman voice. Of course we know the tells. Not only this, but that. You're exactly right! —
The woman and man AI voice endemic to YouTube and ilk is also tremendously off putting. M5Stack has a bunch of these videos, and it devalues what they're doing.
And then... Transformer "art". It is some of the worst drek I've seen. I smell it a mile away. It's easily seen by slop english-like characters. Or too glowy humans. Or overall fake feel. For pixel art, I can perhaps see it. But for anything it just feels... Gross.
I'm completely sure management LOVES it cause its cheap and devalues humans.
> I'm completely sure management LOVES it cause its cheap and devalues humans.
And the tragic bit is that instead of educating them about the pitfalls so everyone's on the same page, so a lot of brave hotheads are literally calling for a class war
> And then... Transformer "art". It is some of the worst drek I've seen. I smell it a mile away.
Hating that every third or fourth ad now is AI-generated. So much worse than what entry-level graphic designers can put out while putting them out of work.
Arts or no arts companies hire you to do X, not to make friends.
If you stop doing X you will be fired/not-hired, simple as.
You can make friends on your free time.
> Arts or no arts companies hire you to do X, not to make friends. If you stop doing X you will be fired/not-hired, simple as.
> You can make friends on your free time.
Most well-adjusted people work to live, they don't live to work. Life comes first, the demands of the job come a distant second.
Great!
All the more reason, then, for them to be okay with not being hired.
Oh my gosh I’ve worked with so many of you!
Go tell HR that you don't give a shit about your coworkers ;)
Spend some time learning about the concept of the excluded middle[1]. It is something that will be useful on your life going forward, it might even save you a lot of trouble.
I am writing a post about that for my blog but it's still not there so ... check back later? Lol.
Anyway, in this particular scenario, a co-worker not being my friend does not mean we are enemies.
1: https://en.wikipedia.org/wiki/Law_of_excluded_middle
Why would this bother HR? They don't give a shit about the coworkers either.
Just to +1 the sentiment here; AI in the gaming industry is absolutely a culture war front. Especially in the indie space, you will lose advertising deals with many sponsors & influencers, for example, if you appear to endorse AI. Probably won't be that way forever, but it's a hotbed right now.
Meanwhile, Asian studios are much more open to experimenting with AI workflows, and will eventually build the next generation of Niantics and AppLovins if this neo-ludditism continues.
Heck, 50% of all Japanese game studios [0] along with all of China's largest studios [1] now use AI within their development pipeline - often with explicit state backing.
You may not like Blizzard or Ubisoft but Tencent, Sony, miHoYo, and even Nintendo are much worse from a work culture, compensation, and work expectation perspective.
[0] - https://www.nikkei.com/article/DGXZQOUF251PU0V20C25A9000000/...
[1] - https://www.scmp.com/tech/tech-trends/article/3341063/next-l...
---
Edit: can't reply
> Why should that be a relevant concern? It's not like any of us will be working in game dev if AI wins.
Becuase AI is not taking all jobs. Yes a lot of redundant work will go away, but there is still a real need for human intervention, monkeypatching, and ingenuity.
The North American gaming industry only exists because the entire ecosystem from AdTech to Engines to Marketplaces exists to develop, finance, and distribute IP.
If you stay frozen in the past, you eventually get outcompeted and the ecosystem will leave. And unlike the automotive industry, game devs aren't a core voting bloc.
This is what happened to the entire animation industry and is what is happening to the film and television industry. Gaming will be the next IP driven industry to leave if everyone remains frozen and opposed to innovation.
> Maybe (sic) Chines sell tools for artists pitching them as useful rather then "haha this will make you obsolete you looser look at slop I made" middle finger pitch?
The obsolence and cost saving message is true though and used all over Asia - even in China [0][1].
Either you innovate and compete, or you will get trounced. THIS is the cultural mindset back in Asia.
Americans best learn how to compete again.
[0] - https://www.zaobao.com.sg/news/china/story20260622-9245522
[1] - https://m.tech.china.com/articles/20260615/202606151894081.h...
"will eventually build the next generation of Niantics and AppLovins"
Why should that be a relevant concern? It's not like any of us will be working in game dev if AI wins.
I dunno, maybe OpenAI and Anthromorphic should learn from Chinese companies how to not act in ridiculously off-putting ways?
Maybe Chines sell tools for artists pitching them as useful rather then "haha this will make you obsolete you looser look at slop I made" middle finger pitch?
> Not everything has to be a culture war front.
This is seemingly spoken from an ignorant and insulated position. The victims of invasion don't get to decide whether or not they live on a war front, nor do the countless skilled and creative individuals losing their entire careers almost overnight.
By the way it's class war you're talking about, not culture war.
This particular technology has been trained by explicitly ripping off a lot of peoples' friends, though.
>Using a technology employers want you to use is not betraying your friends.
If it means helping the employers erase their jobs and their dreams for a career in the industry, then it is.
Propertly testing and reviewing code is a must with AI, it's actually more important. I don't know why the author feels like it's assumed that this goes out the window with AI
Not sure the technology is the issue here. He isn't debating using Java v C#, and one is a betrayal of all that is holy, so he refuses to use it.
I think everyone kind of feels that AI is sucking up all the content that we have all created collectively, and we all know that the bell is tolling for thee, no matter how much you adapt. So if you see friends being fed into a meat grinder, you can have a 'culture war' take on it. He's posting on line to vent, something everyone is venting about.
Of course, for a job, to get a paycheck, we'd use any technology, even if we are the ones running the meat grinder.
I think you might not get the context this is coming from, it’s about a game developer laid off from Blizzard not wanting to partake in the technology that will justify layoffs of their friends. Pretty straight forward to me.
Yeah, but nothing he can do
You will not have a job in software in the future if you refuse to use AI.
I fully stand behind that prediction.
edit: despite the downvotes, I'll double down: most of today's software jobs will disappear. Your job, if it is in software, will disappear. It might transform into something new, if you're lucky. Or it might just go away entirely.
For the vast majority of people who currently have a job in software... you will not have a job in software in the future whether you use AI or not.
The tide is coming for almost all of us.
I guess I'm just confused by this sentiment. Are you making these conclusions while considering the fact that AI is still heavily subsidized? The economics of AI isn't quite the same as other software/tech.
I don't think it's going anywhere, but I don't know what happens when prices start to rise because these companies need to start turning a profit.
I'm happy using GLM 5.2 at API pricing. There will never be a point where essentially unlimited usage (i.e. what I, as a normal developer, can use at peak effort) of extremely good models will become financially unattainable.
> There will never be a point where essentially unlimited usage (i.e. what I, as a normal developer, can use at peak effort) of extremely good models will become financially unattainable.
The way you phrase this prevents refutation. But there will be a point where ordinary individuals cannot participate in the majority of the upside.
In the future, some class of models will require enormous compute that is outside the financial capital capability of ordinary individual contributors, middle class, and upper middle class. This will be sold as a capability to well-funded companies.
Yap, the signs are here.
Software and code is turning into a commodity
> For the vast majority of people who currently have a job in software... you will not have a job in software in the future whether you use AI or not. The tide is coming for almost all of us.
I'll agree with you as soon as all video games are in the metaverse and run natively on the blockchain.
Metaverse is bullshit.
Blockchain is bullshit.
Crypto is bullshit.
VR and wearables are bullshit.
AI does your job. Robots do your job. These are real and substantial and actually provide enormous value. You can get more done per unit of time.
I want them to do my job and everyone else's jobs, I just want to make sure there's a functioning economy we can participate in and benefit from at the other side.
AI makes up shit that doesn't exist and lies to you. Sure, let's get the word salad generator to write all the code. What a great fucking idea.
What year, and how much money are you willing to bet?
Corporations are the organizational equivalent of AI, they one thing well asmass money, their output aside from that is total shit. Not only are you betraying your friends/community/loved ones but such a setup is necessary to prevent the total collapse of society.
yeah, the question is what would you do with $100 million? I mean I don't have $100 million but I'd give it to the arts after I've taken care of myself well and science. Am I gonna make $100 million playing with AI? I don't even know if I can take care of myself, shit, but the government ain't gonna do that. What are the new jobs after the Saaspocalypse? Well all the money went to those companies and what do those companies want or what do those newly minted millionaires want? Hopefully something I have but shit yo. So they're gonna go back to where they came from and be millionaires but how much is Dave Chappelle changed the city that he moved to? Is there a Dave Chappelle center for comedy? some of them want ASI so I look forward to studies on human intelligence. Maybe they'll pay human subjects to model supply two digit numbers while in an MRI machine hopefully enough of those studies to pay my mortgage.
> I mean I don't have $100 million but I'd give it to the arts
I might be naive, but rather than giving money "to the arts", I would much prefer to give money "to the people" en masse, and then leave it up to everyone to decide whether they want to make art for art's sake.
A month ago, I fell back into reading patio11's "don't call yourself a programmer" and I found it fitting. The core of the message wasn't about the title we assign to ourselves but the "other career advice".
I felt compelled to write "don't call yourself a Software engineer" [0], because we are still falling into the same trap of thinking we are hired only for our technical skills.
If we are just looking at a skills and these are assessed by parsing through a resume, then OP is right. We are all at a disadvantage. But the job search starts way before you are looking for a job. It's all about the connections you make along the way.
[0]: https://idiallo.com/blog/you-are-an-ai-enabled-engineer-now
I feel like this advice is not very useful because when you call yourself a software engineer or programmer, you are doing it in order to sell a service.
Your customers are companies looking for someone to slot into a box called "software engineer" and so you sell yourself as such. Nothing wrong with that.
We should also note who Patrick was at the time. He was an SEO consultant and in general a business development expert. It just also happened that he was able to code. And he was very very early to the field. An SEO expert was barely a thing.
So if your only skill is software development, then of course you would call yourself that. And if your main skill is SEO or some other marketing channel, then you call yourself that.
I think the real takeaway from the advice "don't call yourself a programmer" is to search the market for higher paid opportunities, where you can still leverage coding. And you can call yourself a programmer while doing so.
>Engineers in particular are usually very highly paid Cost Centers, which sets MBA’s optimization antennae to twitching. This is what brings us wonderful ideas like outsourcing, which is “Let’s replace really expensive Cost Centers who do some magic which we kinda need but don’t really care about with less expensive Cost Centers in a lower wage country”. (Quick sidenote: You can absolutely ignore outsourcing as a career threat if you read the rest of this guide.) Nobody ever outsources Profit Centers. Attempting to do so would be the setup for MBA humor. It’s like suggesting replacing your source control system with a bunch of copies maintained on floppy disks.
Really good stuff haha.
EDIT: https://www.kalzumeus.com/2011/10/28/dont-call-yourself-a-pr...
Two weeks ago, I got 100/100 of a test from a big company for a first screening without using AI. I was pretty confident that I would pass the first round, even hinted few of my friends, but ended up being rejected with an automated mail… The job market is insane at this point and I am not sure what the recruiters are actually looking for. If the candidate uses AI they’re eliminated, if not they’re eliminated. I guess this is one of these times we read on history books: great unemployement.
Nowadays more than half of job postings are fake. It's either process to show activity or they already have one to hire but need to follow established process.
Getting through an interview process during a bull market has always had a small random component. I think we all have to understand and accept that in bear hiring markets, almost the entire component is random. Having a perfect skill quiz or hackerrank score and getting rejected should not cause you to try to figure out what you did wrong.
You can do everything right in an interview and still get denied. It is truly a crapshoot in today's job market.
> The job market is insane at this point and I am not sure what the recruiters are actually looking for.
To justify their own jobs.
In theory there's never been a better time to hire on SWE talent. There are lots and lots of candidates who rode high during the COVID hiring wave, took on debt based on a high income, got fired, and now need money.
But hiring isn't picking up. You have a bunch of people in the HR industry who realize that for the most part, the combination of candidate filtering, ML, and a basic tech interview process could probably do their jobs. So they have to make the process as byzantine and difficult as possible to be able to go to the suits and say "look at all of these low-quality candidates we kept out!"
> I guess this is one of these times we read on history books: great unemployement.
The good/bad news is that if this continues, there will be either a regression to the mean or a massive de-stablization of most societies. You can't kick most of the working-age population out of their jobs.
I was just involved in the hiring process for a new rec on a team adjacent to mine, and the HR/screening process was a complete mystery as to candidate filtering for outside applicants. So many AI tools making arbitrary judgments from resume content to psychological profiles of candidates, led us to 0 organic people “qualified” enough to interview for an SRE/Devops role. All we hire now is referrals as they are the only ones to get pass automated screening
The industry will realize that while getting LLMs to write code is easy, getting LLMs to write good, production ready code is a skill all on its own, which simply must be done by a human and is not automatable to an LLM in any sense effectively. That will be the differentiating factor software engineering in the future, I think.
If I'm being blunt, if you are in the game industry, you probably have nothing to worry about in terms of LLM coding replacing you, because the tooling used in the gaming industry is as unfriendly to LLM coding as it gets: Heavily visual scripting based, extremely reflection heavy, and the code, Unreal C++ and Unity C#, looks like regular C++/C#, but doesn't behave like normal C++/C#. LLMs simply cannot reason about hidden implicit states effectively, so if the code looks right but doesn't act right, LLMs will simply get confused and start hallucinating.
I've spoken about my troubles getting as much mileage out of LLMs as before while being vague about my industry and, well, this hits the nail in the head.
> because the tooling used in the gaming industry is as unfriendly to LLM coding as it gets: Heavily visual scripting based
At the start of the LLM craze, we (as a team) put ChatGPT to test with Godot. It wasn't very successful in that, IIRC, GDScript 2 was just released and ChatGPT's training corpus was so obviously based on GDScript 1.
We could make ChatGPT outline steps of how to accomplish things in Godot, sort of like getting a customized tutorial. When code isn't involved, ChatGPT was okay and Gemini seemed to fare better. Based on vibes, I think Gemini had a marginally better handle on GDScript 2 for some reason.
I've been trying it lately with Claude still with mixed results. I had to install a few skills/extensions for it (can't tell you which as I basically just blindly followed what our AI advocate recommended). Sometimes it works but when it doesn't it's harder to put a finger on why. Overall I prefer the DX of generating customized tutorials with ChatGPT.
> extremely reflection heavy
Big time. And IME we don't even have to deal with textbook reflection here. Game entities are so convoluted (remember the Fallout 3 Train NPC, the stories of how Skyrim works, etc.) that it is really pushing inheritance and OO to contortions it shouldn't be doing.
Dirty confession: in our game we have this GIANT switch-statement dealing with game objects. It happens in a handful of places, for different game object types. LLMs (Copilot and Codex) could generate the monkey code of adding a switch case and even writing the body but sooner than later, when the new objects have to interact with others, LLMs just can't reason around it. Not to mention the hundred edge cases you have to consider!
And before some smart-ass comments: in my almost-decade of dealing with this code base there have been a handful of attempts to "refactor" these switch-statements, always a newcomer's enthusiastic effort. I'm proud to say, I've managed to slay one of the giants, the only successful effort to my knowledge, and this only happened last year. But I did so by basically delegating to another of the giants; they turned out to be twins and we could do without one. The dirty way is the clean way because the alternative contains Lovecraftian geometries.
Claude has been great for finding edge-case bugs but that's only once the code has been written properly. Generally if QA reports a bug on a pre-release feature, it's at most 50/50 if Claude can debug it. But if it's a player report/live incident, I'd say Claude's chances goes up to around 80%.
All that said,
> if you are in the game industry, you probably have nothing to worry about in terms of LLM coding replacing you
Hah-hah. Look, I agree with you but please get in touch with upper management. As an engineer I'm confident on the value I bring to the table but I'm not sure management "gets" that. Like, no matter how I tell them what features I shipped, what infra I built, etc., it doesn't come across to them that LLMs would not have been able to automate that output!
I am currently quitting a company of 10 years of employment. And I keep hearing how everything's shit. Btw. I'm located in Tokyo where it isn't as bad, apparently, but...
Let's see. My plan:
- Have my own company and start looking for customers. (Rust consulting)
- Keep looking for job opportunities, but don't succumb for shit jobs.
It might be that I'm too hopeful, but you can't know unless you try.
Anyway, I may join the "everything is shit" crowd in half a year if nothing pans out, but until then, I'm hopeful.
why not start looking for customers before quitting?
I suspect the reason is highly personal to their particular situation, but also most likely justified.
Best of luck!
Thank you!
Don't leave unless you have something lined up lol
That might be a truism that's generally applicable, but some people have enough of a safety net to last them several years while building a business. We don't know OP's circumstances.
There are a number of reasons I’d site for the current job market tightness:
- political: there’s an enormous amount of uncertainty here. All businesses make plans and uncertainty puts them all on pause.
- economic: related to political, but we’re teetering on a very bad recession. Watch where national oil reserves go.
- AI: I throw this in with every new technology that comes out. There is always a period of chaos before normalization. We’re still in the chaos phase.
- Business Pain: Right now I don’t see any sector that’s in pain. Inflation has hurt consumers, but we’re still spending. When consumers lock it down, that pain comes back and job market shifts with it.
I have no solution other than figuring out a way to do your own thing. There’s no better time to be a founder.
* cite
> There is always a period of chaos before normalization.
In this case, it's the normalization period that has people terrified.
Understandable. Look at the possible directions:
- GenAI becomes a foundational requirement for tech and non tech sectors. If you’ve refused to engage, you’ve self-selected out of any of those sectors.
- GenAI usage shifts down to just the tech sector, but in an integrated fashion where current engineering practices are still desired. Everyone survives, but pay scales are adjusted down by a not-insignificant amount.
- GenAI bubbles badly, OpenAI and Anthropic merge with Google/Microsoft/Oracle/IBM/???. Tokens become extremely expensive and no one is leaning into agentic integration. Everyone thrives.
The problem with scenario one is it's still cope: "if you didn't develop the skills to use GenAI then you'll be left behind".
But that's not the promise of GenAI models. The skill floor is constantly lowering and your advanced workflow is rendered obsolete monthly.
It might also be foolish, but I'm mid-pivot to becoming an Actuary. They credential through a very transparent exam system, and interviews are relatively cursory, assuming you've passed the appropriate exams on the right timeline. I've got a math degree, and my software experience is all in the data engineering space, which seems to be in demand.
I got laid off from a large company last summer, and took some time off to travel. Now I've got a chill, low paid dev job in a resort town in New Zealand, but my sense is that dev work is not going to be the thing for me long term. This job will pay my bills while I pivot, but they're not going to sponsor me to extend my visa, so I'm on a bit of a timer. The market back in the US seems like an ongoing mess and I don't want to get back into it.
Curious about the perspective from anyone who has a skill set / reputation where finding jobs is easy right now.
I imagine people in ML or who've found a good way to demonstrate prowess with agentic systems may be highly in demand right now(?)
I work in an industry tangentially involved with the ML build-out (think companies like Broadcom, Marvell, etc.). We can't find enough people, if you have like 3+ years of experience with PCIe, Ethernet, DDR, you're a shoe in. Verification, Validation, Design, Customer Applications, Firmware, you name it, we need it. The pay is good too, especially for people who got in a year ago or more, stock base compensation has taken off like a rocket.
Hiring here is a little bit more old school, I guess? Especially because the types of roles we are hiring now are usually 5+ years of experience, we focus more on learning about what the candidates have done in the past, the leetcode type of question interview is just a small part, and matters more for prospective Jr. hires.
That being said, we aren't hiring that many fresh graduates anymore, we already have some we hired, we're focused on investing in them, getting them to learn more about our hardware and code, etc. and hoping to retain them.
If you cannot find enough people in current market then pay is not that good as you think.
I feel like any advice you get from someone would be as useful as "how to be a good coin flipper" advice from the 1 person in 1024 who flipped heads 10 times in a row. In other words it would be purely survivorship bias.
I joined a new company 6 months ago. I interviewed at 16 companies and got 5 offers from a mix of ai cos / big tech / trading firms
Background is SWE at an AI co that's in the news sometimes
It felt about the same in terms of grind effort from my last search in 2022. the main difference was ai companies cared a lot about your understanding of agentic systems and harness / context engineering, and had much more practical rounds with less leetcode (usually 1 medium). More legacy firms (finance / some big tech) still expected you to solve 3-4 leetcode medium/hards throughout the process
Anecdotally I've seen the latter to be true and have had third-party recruiters echo that familiarity with AI use in coding has become a pivotal part of job interviews with startups.
That being said, I'm not sure how much job security having such prowess would convey because I feel AI will be better than us at that too eventually (if not already).
I'm starting a new job in a few weeks, and can confirm (for startups at least) experience with coding agents is something companies are looking for. Multiple companies I interviewed with had a AI assisted interview session to go along with a more typical closed book programming session. I was asked about my use of coding agents in behavioral interviews. I'm not an ML guy, just a generalist SWE with 4 YOE. I only got one offer in my search, but it only took ~a month and I feel pretty good about being able to get more offers with more searching. It helps that I'm young, no dependents, and willing to relocate.
My perspective is it is impossible to cut through the three layers of bullshit between you and anyone who knows what they are talking about. The only way to do this is with brand-name qualifications, like "MIT graduate", not things that are actually impressive. This is also why you see senior developers saying, "the offers I'm getting are bigger and bigger," meanwhile skilled younger developers need to become a marketing professional just to get an interview.
Recruiters have utterly given up on being efficient in the market. I do not know why, but there is something very wrong given "spamming the same brand-name fish all the other recruiters are spamming" is their only strategy. My guess is there is a combination of bad (or an entire lack of) hygienic data filtering and a disconnect between compensation and terminal goals (hiring the best candidates).
What do you mean by easy? Do you mean FAANG or equivalent salary? What level of seniority?
Can speak to my experience that if you are a senior engineer in London the market is relatively easy at the moment (or was at the beginning of the year) even with no connections.
Silicon design/verification. In really high demand at the moment, I guess because of the death of Moore's Law - now it is much more worthwhile making custom chips.
Easy hires are the same now as in the past. They have held several specialized roles before without letting it narrow their career path, and they have at least a decade of experience.
The problem with specialized roles is that nothing lasts too long in software. Given enough time in it, nobody really has an edge. Everyone is smart enough to have invented and implemented the very thing eating the world right now. They just don't have supervillain money or clout, so they work for you instead.
> Curious about the perspective from anyone who has a skill set / reputation where finding jobs is easy right now.
It's easy to find jobs in software engineering provided you have an attractive resume.
What, in your evaluation, are the top two or three attributes of an attractive resume?
There's only one: you work at a top-tier company.
OP's opinion about AI coding is pretty obvious in this blog post. Maybe some of that sips out during interviews which certainly will spook the employer.
In the game industry this is actually a positive thing. If you think that AI can write code better than you can, I'm not going to disagree with you, but maybe you should try getting better or more specialized at what you do, because I can ask the AI simple questions about my field and it's completely wrong 3-4 prompts in.
Using this technology to build anything remotely serious, OR EVEN GAMES, is wildly stupid.
I'm not used to seeing informed takes on the games industry on Hacker news, but then I looked up to see your name. Love your stuff, your website was mind opening when I was in college.
Thank you sir
If you don't have a full time job for a year, why don't you start a software business on your own? It's hard to succeed, but, successful or now, you will have something to show people and something that you are passionate to talk about. It will give you a new way to connect to people, and quite likely will help you to get hired. And who knows, maybe you will succeed.
100% this. Interviewing isn't something that can compound. Striking out from company after company doesn't leave behind a trail of real work and real lessons. Starting a business is tough but it really does teach skills that are hard to find any other way (about sales, recruiting, management, etc). After a certain point, it's wiser to give up on getting hired, and just hire yourself and build something.
I wasn't there but this seems like the same feelings people would have had in the Rust Belt when the first factories started closing and getting a job started getting harder.
During the pandemic money printing things got very weird. It created a lot of leverage and bullshit companies and bullshit dev work which led to artificial demand for software developers.
We are still in the post-pandemic hangover.
If you look up M2 money supply on St Louis fed - that chart has more influence on the job market in the US than anything.
The macro whiplash compounds this problem for people like OP in a few ways:
- cheap money leads to hiring frenzy (cheap capital costs lead to investments in human capital in software)
- developers get conditioned to artificially high demand and assume it will be like that forever
- artificially high demand attract people into software dev for the money instead of love of the art (increasing supply)
- when capital gets expensive again companies have to correct for over-hiring with layoffs and hiring freezes
- developers are stuck in a market with crashing demand (because of higher cost of capital) and over-supply (people attracted to work when cost of capital was cheap)
Everyone says it's about AI, but AI is more like the flavor & scapegoat, the substance is all a consequence of macro policy.
The next time the fed does quant easing labor market will kick up again.
Nobody likes to hear it, but this is the only explanation that makes sense. We had an unprecedented economic shock, and we're dealing with an unprecedented economic fallout. The only question is how much longer it will last.
The timeline also matches up exactly, the Fed started raising rates in Q1-Q3 2022 and you can see the largest spike of layoffs happened Q1-Q3 2022:
https://tradingeconomics.com/united-states/interest-rate
https://layoffs.fyi/
ChatGPT wasn’t released until the end of 2022, and it wasn’t until the next spring where it really started to take off. Spring 2023 is when ChatGPT started to gain traction as a direct to consumer app and it wasn’t until later in the year when startups started building on top of it.
> when I need to recall off the top of my head the proper way to instantiate a list or heap in X language
You only need to know for one though...
It was an example.
I guess even an off hand example, now must be completely 100% technically correct , or you aren't worth a job?
And, I'm pretty sure in "X" language, you can call them differently. Since it is "X", how do you know.
Just use python for all practical programming problems. Lists, sets, and dicts are all you need for most leetcode problems; dynamic typing is convenient; there's good ergonomics for http and other random utility tasks; and pretty much every company is cool with python in an interview. You'll probably only see language trivia questions for languages you claim a specialty in (there's a huge market for C++ specialists, for instance).
> I guess even an off hand example, now must be completely 100% technically correct , or you aren't worth a job?
It's a litmus test, and not a terribly challenging one. It's solved by spending a week doing simple coding challenges.
> This is probably the worst job market I've seen in a while.
What a noncommittal sentence
I noted this too - considering he's only been a developer for a decade, this is a weird comment to make - the last decade has been a breeze.
Adapt dude. You can.
Everyone says Adapt.
Like, go be a farmer, Adapt? Reinvent yourself as a performance artist? Because, learn Java in 21 days, is kind of gone.
The blog lacks crucial info. What type of projects have you worked on? What work are you applying for? Personally, "small contractor" and Blizzard does not translate well to the typical "enterprise web dev" role.
I really hope this thread is wrong.
In 5 years, the Junior pipeline will be completely dry.
Seniors will be retiring.
Companies will be floundering.
We'll see a great correction where we need workers again.
Programmers/SE/etc... will be needed again. Always were needed, but at least managers will realize it again.
The companies seem to truly believe they will just have a handful of devs at that point manning large pools of AI agents. I doubt it though!
The first place to look for jobs should be in your network, people that worked with you, teachers, ex-managers.
Applying for jobs out of the blue usually sucks. In the ideal world, you want recruiters calling out to you.
Don't assume you can't do proper software engineering using AI. You can. The people that want to create loops are not the only ones delivering with acceptable productivity. Lot's of us still write code, at least interfaces, traits, modules or whatever, and just use the AI to fill the blanks on the really tedious code.
The job market to be honest has been very fucked. To me a lot of this sounds like people experiencing how terrible tech hiring has become for the first time after being in a stable job for a long time. Almost everything the Author said, was something I’ve experienced when I was laid off in the 2022-2023 wave of layoffs. At the time I was told “it was a skill issue”.
Hopefully you had been saving and investing folks. The sun has set, the power is off, the signal is lost. See you on the other side!
It really is awful right now. I'm lucky enough to still have a job, but floated my resume around earlier this year. I have a pretty good resume and and 15 YOE, and got turned down EVERYWHERE. I used to at least get interviews at like 50% of places I applied to.
And then recruiter spam is COMPLETELY gone. I'm not really complaining, but it feels indicative of where things are at.
> And then recruiter spam is COMPLETELY gone.
Oddly enough, in the past year recruiter spam has ticked up significantly for me. It was completely gone for a while, but it's back in full force.
You're right on the latter point. I always wondered if Linkedin increased their spam filter or make it 10x more expensive to DM candidates, but I guess it's simply because there is simply a lack of demand right now
it is quite dystopic
Companies should encourage AI use in interviews to avoid this issue.
agree with that.
Freelance? Find someone who needs or wants a thing done, but doesn't know how to do it, then do it for them. Take half of the money upfront.
I'm transitioning to freelance (independent consultant, full stack and Cloud), and I find I'm working alongside AI in business manager's hands. It's an uncanny mode of competition.
most of those who might hire someone because they don't know how to do it are now using AI instead. so that market is shrinking too.
Related:
The early hiring funnel is now breaking on both ends
https://news.ycombinator.com/item?id=48620142
I agree that the online hackerrank quizes where it isn't even a video call is dumb because so many people cheat and if you don't, you're at a disadvantage.
Lots I agree with here, but...
> I would promptly remove them from my LinkedIn connections afterwards because I'm exhausted of pointless connections and recruiters.
Why would you do something like this, it's just counter productive. I've had numerous recruiters reach out weeks or months later to say "hey another team is interested", or even when they have moved on to other jobs.
Stop being so bitter you're just shooting yourself in the foot.
I'd suggest not adding recruiters as LinkedIn connections in the first place. I mean, LinkedIn is hardly sacred, but why are you adding people you don't know as a connection?
Recruiters only add you as a connection if they can't afford LinkedIn premium, which is what you need to message people you're not connected to (except for connection requests). That probably means that they're not very successful recruiters.
I heard that recruiters also use amount of connections to filter out candidates, and they probably view it as if you don't have 500-1000 connection probably you just don't fit basic requirements of the the role even with 10+ yoe.
Even is so stupid but looks like in last few years lot of strange metrics like that used more and more.
I'm talking about recruiters that you talked to throughout an interview loop, not randos.
Can we please stop highlighting and indulging these neurotic cybersocialists?
If you truly give a shit you have to change and help make the mess less mess. It sucks, it might be worse than it was, but you can't continue giving a shit by not participating. The horse has left the barn on this one.
The frontier model companies could all collapse tomorrow but the tech is not going anywhere.
This is peripheral about bearblog, but it's so grating to see the "D M, Y" date format with the comma. The correct format is "D M Y." It's like someone deciding to write June, 6, 2026 for some reason.