Hacker News Reader

Comment by woutr_be | Hacker News Reader

I’m currently doing a project with someone who only uses LLMs, and it’s exhausting and mentally draining.

Whenever I give feedback on something, the answer is just “let me tell Claude”. The person has no understanding of how everything works, and most of the code reflects that.

The other day he hardcoded in a demo mode, simply because he didn’t even know how to set up a local environment and set environment variables. I’m confused as to why Claude didn’t even knew this, but it might just be the prompting.

I limit LLM usage myself, and if I do use it, I try to use it on extremely specific tasks. It’s the only way it works for me.

I honestly don’t understand how all these companies are getting away with generating AI code. Even in a small project I quickly fall behind on my understanding of the project.

View on HN

I remember working with people like this before AI and it was annoying but they struggled with productivity because they didn't understand what they were working on enough to produce good code efficiently, so the problem usually took care of itself.

Now these people can thrive because LLM coding encourages the incurious and punishes the deep thinker.

At least to me, before LLMs, those people could somewhat be trained. You could give them feedback, and they would improve and learn the code.

Now your feedback is just another prompt for them, the code might be slightly better, but the person learned nothing from it.

It punishes the inflexible deep thinker, perhaps. If you can’t figure out how to use an incredibly powerful tool to your advantage, how deeply are you thinking, really?

Exactly. Add it to the fact that the world was never particularly "fair" to deep thinkers, because said deep thinkers are often not prioritizing "production". It already rewarded people who could search for and ship a working solution before they understand it perfectly. The person who can only move after fully understanding the problem was at a disadvantage long before LLMs, almost everywhere outside academia.

Yes, and you can see this principle clearly reflected in the large majority of all software today, which is absolute, dysfunctional, crap.

Well most experienced developers I know hate it and agree it's taken all pleasure out of the work. We've spent our careers designing and building high quality software systems. But now we're told our job is to use the plagiarism box to sling slop all the time. Those who can are transitioning out of the field.

This is an industry where American developers have successfully competed on quality with the rest of the world for years. We never were very cheap but we always were the best and worth our premium. Now that's being destroyed by a short-sighted industry.

I'd like to just do something else and work on open source. Except I know if I contribute to open source my work will just be stolen by the plagiarism bot.

You can read many posts here by very experienced developers who disagree.

It’s possible that the experienced developers you know just aren’t that good at adapting to major changes like this.

Another possibility is that people are afraid to speak up when the threat is termination

"Stroustrup points out that some senior developers are starting to retire because they're tired of constantly validating unpredictable AI code. "

https://www.newsbytesapp.com/news/science/bjarne-stroustrup-...

I'm the same as you. Very specific tasks. Just now I spent two hours in a conversation with claude code to change 5 lines of code that change a significant semantic.

On the other hand my buddy is spending $10k in tokens a day on agents to build something. He's a very smart guy, former developer so it's not just AI psychosis talking.

Still trying to figure it out. Not that I have $10k to spend.

What on earth is he doing with $2.6M in token burn per year? That's at the level where he is running multiple simultaneous frontier models 24/7.

Oh my god, is that in USD?

> $10k in tokens a day

What on earth is he building?

And how can I become so rich?

LLM generated code is obtuse and dense, unnecessarily so

Reply on news.ycombinator.com