If the bubble bursts as hard as it's looking to pop, you may not have to worry too much. It will still be here, but the heavy computing will be very expensive, and what's left will be the assistant type tools. Those are fine. The bubble will burst, it's just a matter of how long it will take. They are very over-leveraged.

https://www.youtube.com/watch?v=3sDzlv-VI2U

Always

Be

Constantly learning

(regardless of trends)

The tech debt that these assholes are generating is either going to cause collapse of the companies and systems, or they're going to have to be fixed. Plan for both.

I'm currently expecting to lose my job over this in the comming months. I'm planning to go to school for a different field. I don't want to deal with tech any more.

Yyyyyup. Lost my last job over AI bullshit, new job uses it even more. It’s inescapable.

I do have the one benefit of being in a lower tier position now so shit breaking from AI use is not my problem to fix, so ̄\(ツ)/ ̄

Remember when one of your professors or mentors told you "Learn Fortran, all the old guys are dying and leaving desperate employers behind."

What do you think happens in five years when no one is willing to pay to train models, the existing ones are no longer keeping up, and most of the industry has spent their time with a tool that actively declines their skills?

I'd like to think that's at the end of this tunnel is a light that means the culling of bootcamp skids and should-be-MBA techbros from the jobs that should be for developers who are in it for the love of the game.

In the meantime? Yeah, its gonna be rough, sis.

I hear you. I am in a better position but I strongly feel the push to use AI at all costs. I hate it. I am mulling over it, and I am on the way to accept being a third rate academic just to keep not using AI

I feel ya. I'm a security tester so I get to face AI daily as well at my work.

Something depressing I've noted as well since the AI age is the (seemingly) complete death of curiosity. Some of the smartest people I know, now completely relegate themselves to become a manager to AI without even touching testing targets anymore.

There used to be a ravenous hunger for new, weird vulnerabilities and quirks in software that could be leveraged. Now they're all just gloating about how they're doing everything talking to a machine over discord.

I feel like the only weirdo that apparently actually liked doing the work. It's sad.

As a grey beard, I never touch AI. My job knows it too. My manager accepts it but tells me to keep it quiet.

They put up with me because I spend 90 % of my current days helping the juniors identify and fix flaws from their AI generated code. The company pushes so hard to use AI, but there's always something to fix and the juniors will never get that needed exposure for troubleshooting code.

🫂

I hear you OP. I am a senior developer and hate my current employer. He is a miserly c*nt. Half of our people already left for better paying jobs. To make matters worse, he has recently joined the Claude cult. I am stranded in search of a new job, that does not include AI.

Our former junior has altogether left development, as there are hardly any junior vacancies anymore. He is flipping burgers now and earning better than he did at our company. He is replaced by Claude by my boss, so one salary less to pay.

As for myself, I can do one of two things:

• Find a new development job. It WILL include AI, which I do not want.
• Leave development entirely. At my age, almost every application is met with utter silence, even for related jobs for which I do have some experience.

I will be opening a very expensive bottle as soon as the bubble bursts.

Yep, it sucks. I feel for you. Now I need to get back to slopping it up at my job.

A lot of folks are saying they wouldn't use Gen-AI even if they have to starve. Don't listen to them. It's better to hate your job and have your basic needs met than to be literally housing/food insecure. If there's blame to be doled (and there is blame to be doled), it's on your boss, his boss, Sam Fucking Altman, and all them. Don't feel bad about taking care of yourself however you need to, even if it means using and making slop. I don't envy you having to make that choice (and I may be in the same position soon), but don't feel bad about making it.

Meanwhile, others who have said the bubble will burst in time are absolutely correct. If you do have to have a job that requires Gen-AI usage, understand it's just to have a paycheck while you wait out the bubble.

None of the above is to say you're overreacting. This is beyond fucked and you have every right to be super pissed. I hope things work out as well as can be expected for you.

i'm trying to find people to work on human-only projects with me. i'm working on a mod manager for the steam frame currently. the AGENTS.md in the repo requires bots to include a phrase in their PRs and issues, so they can be easily filtered and the users banned.

Yup. Even as a mid-career, HIGHLY successful individual, I'm in the same boat. I am so sorry this is happening to you. Every time I feel sorry for myself I think about what the outlook is for those of you just graduating and man, it's fucking bleak. At the end of the day I think we all have to just do it and hope and push for better.

Personally I'm taking advantage of work from home. If they're going to make me use these fucked up bots, fuck em. I'll set the thing running and work on a side project, or play guitar. I refuse to sit there and just stress myself out thinking about my abhorrence for the thing.

Yep, pretty much nailed it. I've been a developer for nearly 15 years now and I haven't enjoyed the introduction of AI into the work flow. It's nice for some things, but my company seems to be pushing toward AI doing all the coding and the "developers" just being reviewers. No on ehas ever told me they like reviewing code. People enjoy the coding part and they want to use AI to take that away.

I've wondered what I'll do if I get to a point where I no longer enjoy my job. Because my company isn't unique in its use of AI. It's going to be every software company.

Plot twist: we use AI and accelerate the downfall of our agricultural yields due to increased data centers (aka blow the climate crisis by so bad). You might starve either way.

Hey mate! My recommendation is that you connect with humanist communities. For example, I'm at a forum called humancrafted. Connection and community is how we stay healthy in these trying times.

It's not always going to be like this. I genuinely think the AI bubble is going to pop and the whole way AI is viewed will change.

I was there for the dotcom bubble. When it popped the Internet didn't go away, it just settled into the tool it was designed to be.

LLMs won't be uninvented... they'll just become tools.

I’m in tech and I think genai / llm is garbage that and its creators should suffer in Hell for all eternity. I’ve only touched it to show how shit it is, and for prompt injection attacks.

I feel this. I'm looking for a job, and in the mean time I'm creating projects for my protfolio where I have to use proprietary crap. I use the AI model, the software company provided because I don't enjoy using that software, but will have to for a living.

What I'm doing to keep my sanity is actually learning another language and contributing to a few projects that have banned LLMs. There are really good projects with good intentions and amazing maintainers who actually need help. They could use our help.

These are dark times, but I believe many meat proxies, and slop will end when the bubble pops. Not to bring US politics into this (I'm not even American) but this recent election result is going to make things difficult for those fucking companies to slither out of. Most of these companies (atleast in my country) are relying on the subsidized plans. When that ends/gets worse (which they will as AI agents are just subscription products), most of these deskilled losers might just move on to the next grift. Many won't even bother or have the hardware (most people in my country can barely afford a PC) to run good models locally

I have hope. We lose only when we give up, Delusional ? Maybe

I've worked in municipal IT my whole career and even during the latest big conference I attended the AI presentations were fairly tame and cautious.

Municipalities move SLOWW and most still don't have a fully fledged out AI policy so no-one is forcing you to use LLMs. I do keep having to tell people that the ai answer they copy pasted to me is completely wrong so that's fun lol.

Is is also nice to know that at the end of the day you're helping someone maintain a city more efficiently for the people living there instead of enriching some corpo overlords.

Pay is ~15% less than industry average but the benefits tend to be good with great job security and hours.

Fr. I have no idea what's the point. It doesn't even help you.

I've avoided taking management roles all my life as I suck at it. AI took away the technical part of my job. I must say it is faster than me but really, I'm gonna go away do something interresting.

Once had a IT job 15 years ago, but the growing uncertainty had me trying out learning the trades.

I left the tech industry after 10 years because of AI

Maybe subscribe to effective altruism *. Become the best slop hardware engineer out there, make a lot of money. And then use that money to fight AI. /s

Yeah. I feel this so much. Just started uni and while I don't use it there, the push from profs is wild too. The worst part is knowing that I will have to use it at work eventually, and that I'm very prone to behavioral addictions so I'm like. Hahah. Scared as hell :/ Unsure what to do anymore.

I haven't used AI at Uni yet, and I'm being kind of cautiously optimistic that the course I have right now where they allow using AI for software development teaches me something about it in a non-fucked up way.

Yeah, it's the new reality. Now not using AI is uneconomical. The job is more focused on security, architecture, design, interacting with clients and a lot of code reviewing.

The value you provide is being a gatekeeper, testing, QA, performance, maintainability and avoiding spaghetti. It's a fundamentally different job and I think the learning curve has shifted a lot. Now the initial learning curve is very gentle but getting to the next level after vibe coding is a very steep curve because knowing what's good code is hard and before came with tons of exposure to writing it.

LLMs behave much better in clear simple easy to reason about codebases but spaghetti side effect codebases cause a lot of trouble. You can still absolutely write yourself into a corner with an LLM.

Yep, the only way I can keep up with the slop is to also slop. It's always been about pushing shitty code quickly, but now nobody gives a damn about code quality.

You're a university student. If you think writing code is the job of an engineer then that is the problem. That was never our job. It was just the main tool we had to use.

Or work in an industry that expects to survive the next 100 years. None of them use AI at all.

I know nothing about coding but surely there are degrees to AI's use as a tool? I've heard it can produce slop but is it always necessarily so? I've heard people using it for debugging, that seems reasonable to someone with my level of knowledge.

I’m not in a computer science or software engineering field. I have been coding for a long time, started with turbo pascal, if that tells you something. AI (LLMs) is horrible for many things, especially writing things for humans, or generating “answers” for people’s questions about the real world.
AI agents are very good at writing code. If a human needs code for work, they are better off telling a well configured and prompted AI what they want done and how, than writing code themselves. There may be certain exceptions, mostly at the more sophisticated levels of performance-critical systems, but even then, AI will make the end product better by helping the human developer by writing and running tests, testing alternative approaches, etc. , because with AI assistance, the dev can simply get more of that done within the given timeframe.
That is all to say, I can’t imagine a job that involves writing code that doesn’t massively benefit from AI. It’s a tool that should be used. You should learn how to use it well, it’s part of your trade. Like you don’t hand punch cards to run machine code. You use an IDE, code completion, version control, linter, etc. You learn these tools as part of your education, because if you don’t use them well, you’re not good at your job. Of course if you code for hobby or whatever, this doesn’t apply.
The comments about after the crash… Agents for writing code will stay after the crash, like the internet stayed after the dotcom bust. Like rn I pay 20 bucks a month for claude, for work. That’s unsustainable, it will and it should cost more. If it costed $2000 a month, I would still keep the sub, without any doubt. Just for comparison, 3 years ago I got a quote from a data science consulting company to maintain a specific project for 30k a year- that didn’t include developing it. A single Claude pro sub is objectively better for us than what that contract offered.