Dejavu, Complaints on using AI sounds very similar to early times of offshoring/outsourcing. At the end of the day, corporates go for most profitable solution, so AI is going to replace some percentage of headcount.
Firing programmers for AI is a mistake
491–500 of 886 posts
Re: Firing programmers for AI is a mistake
#492Earlier quoted context omitted.
> more sophisticated tech stacks Please don't do this, pick more boring tech stacks https://news.ycombinator.com/item?id=43012862 instead. "Sophisticated" tech stacks are a huge waste, so please save the sophisticated stuff for the 0.1% of the time where you actually need it.
That's great advice when you're building a simple CRUD app - use the paved roads for the 10^9th instance. It's terrible advice when you're building something that will cause that boring tech to fall over. Or when you've reached the limits of that boring tech and are still growing. Or when the sophisticated tech lowers CPU usage by 1% and saves your company millions of dollars. Or when that sophisticate tech saves you…
"Boring" tech stacks tend to be highly scalable in their own right - certainly more so than the average of trendy newfangled tech. So what's a lot more likely is that the trendy newfangled tech will fail to meet your needs and you'll be moving to some even newer and trendier tech, at surprisingly high cost. The point of picking the "boring" choice is that it keeps you off that treadmill.
Re: Firing programmers for AI is a mistake
#493Earlier quoted context omitted.
> Why wouldn't this be the case for people using LLM like it was for everyone else? I feel like it's a bit different this time because LLMs aren't just an abstraction. To make an analogy: Ruby on Rails serves a similar role as highways—it's a quick path to get where you're going, but once you learn the major highways in a metro area you can very easily break out and explore and learn the surface streets. LLMs are a G…
> LLMs are a GPS, not a highway. I love these analogies and I think this one is apt. To adapt another which I saw here to this RoR thread, if you're building furniture then LLMs are powertools while frameworks are ikea flatpacks.
It’s best the analogy of the month. I don’t think cab drivers today are the same cab drivers that knew the city by heart of the past.
So, it’s been a privilege gentlemen, writing apps from scratch with you.
walks off the programming Titanic with a giant violin
Re: Firing programmers for AI is a mistake
#494There's such a huge disconnect between people reading headlines and developers who are actually trying to use AI day to day in good faith. We know what it is good at and what it's not. It's incredibly far away from doing any significant change in a mature codebase. In fact I've become so bearish on the technology trying to use it for this, I'm thinking there's going to have to be some other breakthrough or something…
I think that LLMs are only going to make people with real tech/programming skills much more in demand, as younger programmers skip straight into prompt engineering and never develop themselves technically beyond the bare minimum needed to glue things together. The gap between people with deep, hands-on experience that understand how a computer works and prompt engineers will become so insanely deep. Somebody needs to…
I don’t think “prompt engineering” will remain its own field.
It’s just modern SEO and SEO will eat it, eventually.
Prompt engineering as a service makes more sense than having on-staff people anyway, since your prompt’s effectiveness can change from model to model.
Have someone else deal with platform inconsistencies, like always.
Re: Firing programmers for AI is a mistake
#495There's such a huge disconnect between people reading headlines and developers who are actually trying to use AI day to day in good faith. We know what it is good at and what it's not. It's incredibly far away from doing any significant change in a mature codebase. In fact I've become so bearish on the technology trying to use it for this, I'm thinking there's going to have to be some other breakthrough or something…
A friend of mine reached out with some code ChatGPT wrote for him to trade crypto. It had so much random crap in it and lines would say "AI enhanced trading algo" and it was just an np.randomint line. It was pulling in random deps not even used. I get it though like I'm terrible working with IMUs and I want to just get something going but I can't there's that wall I need to overcome/learn eg. the math behind it. Same…
I used claude to help write a crypto trading bot. It helped me push out thousands of lines a day. What wouldve taken months took a couple weeks. Obviously you still need experienced pilots but unless we find an absolute fuckload of new work to do(not unlikely looking at history) its hard for me to see anything other than way less developers being needed.
Re: Firing programmers for AI is a mistake
#496Earlier quoted context omitted.
There is little evidence that AI is replacing engineers, but there is a whole lot of evidence that shareholders and execs really love the idea and are trying every angle to achieve it.
What's the far future end-state that these shareholders and execs envision? Companies with no staff? Just self-maintaining robots in the factory and AI doing the office jobs and paperwork? And a single CEO sitting in a chair prompting them all? Is that what shareholders see as the future of business? Who has money to buy the company's products? Other CEOs?
Just a paperclip maximizer, with all humans reduced to shareholders in the paperclip maximizer, and also possibly future paperclips.
Re: Firing programmers for AI is a mistake
#497Earlier quoted context omitted.
I think also a critical weakness is that LLMs are trained on the code people write ... and our code doesn't annotate what was written by a human and what was suggested by a tool. In your analogy, this would be like if your sat nav system suggests that you turn right where other people have turned right ... because they were directed to turn by their sat nav.
That already happens. Maps directs you to odd nonsense detours somewhat frequently now, that you get better results by overriding the machine. It's going down the way of web search.
> It's going down the way of web search.
This is an interesting idea. There's an obvious force directing search to get worse, which is the adversarial desire of certain people to be found.
But no such force exists for directions. Why would they be getting worse?
Re: Firing programmers for AI is a mistake
#498Earlier quoted context omitted.
Not to mention I haven't really seen AI replace anyone, except perhaps as a scapegoat for execs who were planning on layoffs anyway. That said, I do think there is real risk of letting AI hinder the growth of Junior dev talent.
I think I've seen us be able to do more with fewer people than in the past. But that isn't the limiting factor for our hiring. All else equal, we'd like to just do more , when we can afford to hire the people. There isn't a fixed amount of work to be done. We have lots of ideas for products and services to make if we have the capacity.
Agreed, I often see AI discussed as if most companies wouldn't take 10x more developers if they could have them for free
Re: Firing programmers for AI is a mistake
#499Agreed - AI agents are simply a form of outsourcing and companies that go all in on that premise will get bonked hard.
Getting into cyber security might be a gold mine in spite of all the AI generated code that is going to be churned out in this transition period.
Re: Firing programmers for AI is a mistake
#500Earlier quoted context omitted.
>Of course both these assumptions are wrong, the quality of code produced by AI will improve dramatically as model evolves. How are you so sure?
I'm not sure, it's an observation considering how AI improvement is related to Moore's law. [1]( https://techcrunch.com/2025/01/07/nvidia-ceo-says-his-ai-chi... )
However, an increase in computing quality doesn't necessarily mean an increase in output quality, as you need compute power + data to train these models.
Just increasing compute power will increase the performance/training speed of these models, but you also need to increase the quality of the data that you are training these models on.
Maybe... the reason why these models show a high school level of understanding is because most of the data on the internet that these models have been trained on is of high school graduate quality.