I fired my entire dev team. Replaced them with O1, Lovable, and Cursor. Now I ship 100X faster with code that’s 10X cleaner. Open AI o3 is coming, and 90% of dev jobs won’t survive.
Managers who try to replace engineers with AI are probably shitty bosses to work for anyway, because they don’t have a single fucking clue about what the job actually entails.
The only thing those employees lost was their income. I’d move on and never look back.
Managers who try to replace engineers with AI are probably shitty bosses to work for anyway, because they don’t have a single fucking clue about what the job actually entails.
Replace “AI” with anything or anyone promising quick, easy or cheap solutions.
This decade it’s AI, last decade it was outsourcing to the third world. The assault on skilled workers never stops.
Skilled workers cost money.
Shareholders hate that.
*last four decades
Yeah I remember when I started my career 15 years ago they were already trying to re-internalize projects after having horrible off -shore experience…
Arguably it’s been going on since Nixon normalized relations with China, which led to the factories moving there for cheaper labor and the eventual hollowing out of the so called middle class in America.
It really took off in the 90s though with the passage of NAFTA and the establishment of the idea of globalization and neoliberalism aka “third way” thinking and the idea of the “end of history” (Francis Fukuyama) after the fall of the USSR. (He has since reneged on a lot of ideas that made him popular in the first place).
It most certainly has not been just a decade though. Any fair argument is gonna pin it on being the last 30-50 or so years.
Few managers have a clue about what is needed or how, the good ones knows that, the bad thinks things like “it can’t be that hard!”.
I had a manager tell my peer, “that guy’s job, I could Google for like 2 months and totally do all he does.”
Yeah, he wasn’t my manager for long.
Wow. Like lol, sure!
Probably projection: “anyone can google for 2 months and do my job”
Honestly, it’s ridiculous to me, how much bullshit you can get peddled, if you’re ready to eat it. Like, absolutely fuck the guy for being proud of firing his workforce, but I cannot blame him for believing the bullshit. At this point, you can readily find scientific studies claiming that AI surpasses humans, and if you look into them, they’re just using a bad test setup. Because well, guess what gets you into the news. It’s certainly not “AI is only reliably applicable for certain niche use-cases and needs guidance from someone with real intelligence”.
but I cannot blame him for believing the bullshit
Sure you can! I do.
but I cannot blame him for believing the bullshit.
I can. Managers are supposed to have at least a basic understanding of the thing they’re in charge of. If the manager just follows online hype, they’re no better than AI.
You’re describing half of middle managers right now
This is the sort of attitude I’d expect from people who constantly fall for bullshit.
Where is the responsibility?
I just wish I had lower standards and was willing to take advantage of people like this. I’d be rich.
The important thing about lowering your standards is that you need to keep your skepticism otherwise you’re going to end up firing your entire workforce and then replacing them with non-functional AI.
they’re just using a bad test setup
On a meta level, it is pretty funny to think about. “Software that was trained to know something, can recite that knowledge back.”
I’ve heard of these mystical paper tomes with knowledge in them, flip to the right page, and it will show you how to perform rocket surgery. It knows! It must be intelligent! What amazing times we live in!
Remember the dot com bust? Well, so do these people. They’re not stupid.
Missing out on the AI revolution can mean the death of your business and the possible rewards are astronomical. These guys are well aware of the risks. They’re aiming to be the 1% that comes out on top after the storm is past.
Also this: Imagine you’re a CEO and the board confronts you, “How are you integrating AI into our business model?” You better have an answer or you’re fired in favor of a new CEO who will engage with AI.
as someone else very recently posted elsewhere on lemmy:
https://lemmy.world/pictrs/image/c81a4a7c-865c-4250-b89c-96e0105af62e.jpeg
A small part of me hopes this was AI generated.
as amusing as the irony would be, it’s clearly too clean, sharp, organized, and consistent to be AI.
Fancy being so completely confident that you didn’t even test the AI before firing your entire human workforce.
What’s a prat
As much as I want to believe this, they didn’t even link to the referenced Reddit post.
0/10 - Gave me hope and dashed it, -1/10 with rice
Naturally, it’s all over programming.dev.
Here’s one of the links there: https://slrpnk.net/pictrs/image/d4cd237a-0ad7-4b9c-9548-c34c63194164.jpeg?thumbnail=1500&format=webp
And
I don’t see any evidence this actually happened besides some news article talking about some guy’s tweets. Given the hyperbole, I’m more likely to think he’s just lying through his teeth and didn’t do any of those. I can tweet whatever the hell I want and pretend to own a company, too.
Open AI o3 is coming, and 90% of dev jobs won’t survive.
Developers will be the ones shutting the door on the way out to the singularity event.
a hallucination, perhaps
Another Lemmy post read:
As much as I want to believe this, they didn’t even link to the referenced Reddit post.
Sorry - I would have looked for a better source if this was a real news community.
This… Is a real news community? The whole point is that it’s not The Onion?
I mean, it’s not not not the Onion.
This is sort of what I have been saying.
It’s not that AI is capable of taking jobs yet, but if corporate management doesn’t understand that, it really doesn’t matter. They’ll fire most of the team and tell the remaining few people to make it work.
Later: “No one could have predicted this! Bailout please.”
The thing is, a lot more useful AI tools could be developed if they didn’t try to actively replace the workers, but to create tools for them.
With Neural Amp Modeler, I can simulate an expensive tube amp without owning it. However, I still need to play my guitar, I cannot tell my computer “hey PC, play some really sick, heavy, and bluesy sludge metal riff” instead.
People still mistakenly believe that LLMs can think. Writing code that compiles is the easiest part of being a developer.
Best it can do is to speed up some developer’s output, which is extremely dependent on what language they use, their target application, etc. Sometimes it has the opposite effect.
I agree.
As someone who uses a number of LLMs often as a pair programmer / sounding board - they’re incredibly useful if you have a very clear idea of your goal and also a solid idea of the architectural patterns you’re going after because they’re so often flatly wrong or suggest solutions that are wildly inefficient/inappropriate for larger projects/applications.
The more context you can provide the better it does but it still falls over often - suggesting courses of action or solutions that are completely hallucinated.
The one thing that’s consistently true is that the better I can describe my goals the better the response tends to be.
The best part about using them is that, for the most challenging work, I find that forcing myself to clearly explain my problem and goals in writing often leads me to a solution without ever having to submit a request.
There’s something about trying to clearly explain the problem to someone who “doesn’t know the space” that’s been helpful to finding the solution.
It’s odd that I had to rediscover this in such a visceral way after a previous life as a tech product person.
Let’s suppose they somehow perfect LLMs to write code.
So now we can tell in plain English what our applications need to do! But English (and most human languages) are open to interpretation so our prompts need to be really detailed and unequivocal. Hum, wasn’t that the reason we made programming languages in the first place? In order to provide concise and unequivocal instructions to computers?
And as you say, LLMs don’t think, don’t have domain-specific knowledge they can apply, and don’t ask questions where things are unclear or problematic.
Crappy copy/pasta jobs for crappy applications will suffer but there will always be a place for talented programmers.
That website is cancer AF. Did they replace the entire staff at techgig with AI or something?
More (not The Onion) sources:
