Hacker Newsnew | past | comments | ask | show | jobs | submit | Kiro's commentslogin

Failing 9 out of 10 times for such simple tasks is indeed puzzling. I have no idea what you're doing to achieve that but I'm impressed.

> Crafted like poetry

I find this idea that humans all of a sudden write beautiful code very funny. Most code produced by hand is dirty and filled with ugly hacks. The argument might work against AI art but falls flat for programming.


It's not cheap but it's also not unusual for devs to burn $200 a day on tokens.

The best programmers I know are the ones most excited about it.

The mediocre programmers who are toxic gate keepers seem to be the ones most upset by it.


Yeah right. Only mediocre people like Rob Pike would be a toxic gate keeper.

The reality is that in the theft of Chardet at least 2000 people supported Mark Pilgrim and almost no one supported the three programmers who constantly blog about AI and try to reprogram people.

Incidentally, everyone who unironically uses the word "gate keeper" is mediocre.


definitely. With AI I can stop working on the painful tasks and spend much more time on things that matter most to me: building the right abstractions, thinking about the maths, talking to the customer...

But TBH, I have been a bit "shocked" by AI as well. It's much more troubling that the coming of the internet. But my hope is that having worked with AI extensively for the past 1-2 years, I'm confident they miss the important things: how to build the abstractions to solve the non-code constraints (like ease of maintenance, explainability to others, etc.)

And the way it goes at the moment shows no sign of progress in that area (throwing more agents at a problem will not help).


I like that style. It's a very efficient way to convey information and ideas. Reposting it as your own text however is obviously not a good idea since it's so easy to recognize.


> If I could give any advice to someone who needs it, I’d tell them to write down the things that have made them happy, and then explore why.

> But I really hope to live in a world where my future kids find sitting in front of a rectangle all day to be dystopian and cringe.

What if sitting in front of a rectangle is the thing that makes you happy?


And yet people are using it for that, even if it's not rational. I use ChatGPT for some things that would be easier and better to do with other tools out of habit.


> but I've never seen it develop something more than trivial correctly.

What are you working on? I personally haven't seen LLMs struggle with any kind of problem in months. Legacy codebase with great complexity and performance-critical code. No issue whatsoever regardless of the size of the task.


LLMs are best at JavaScript.


Which happens to be the majority of code they had stolen.


Yes, so Electron is naturally the best target for any LLM.


I personally love that phrasing even if it's a clear tell. Comparisons work well for me to grasp an idea. I also love bullet points.

So yeah, I guess I like LLM writing.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: