I've been posting recently how I refactored a few different code bases with the help of AI. Faster code, higher quality code, overall smaller. AI is not a hammer, it's a Lathe: incredibly powerful but only if you understand exactly what you're doing otherwise it will happily make a big mess.
if you have to understand exactly what you're doing, why not just... do it?
That question completely misunderstands what AI is for. Why would I just do it when the AI did it for me in less time that I could myself and mechanically in a way that is arguably harder for a human to do? AI is surprisingly good at identifying all the edge cases.
i probably don't understand. the main thought i have re: llm coding is, why i would want to talk to a insipid, pandering chatbot instead of having fun writing code?
but, as an engineer, i have to say if it works for you and you're getting quality output, then go for it. it's just not for me.
It seems to me you're coming in with a negative preconceptions (e.g. "insipid, pandering chatbot"). What part about coding is fun for you? What part is boring? Keep the fun bits and take the boring bits and have the LLM do those.
> Faster code, higher quality code, overall smaller.
I'll have to take your word for it, I have yet to see a PR that used AI that wasn't slop.
> AI is not a hammer, it's a Lathe
I would liken it more to dynamite.
> I'll have to take your word for it, I have yet to see a PR that used AI that wasn't slop.
How would you know a non-slop PR didn't use AI?
Why would I accept slop out of the AI? I don't. So I don't have any.
I don't understand the disconnect here. Some people really want to be extremely negative about this pretty amazing technology while the rest of us have just incorporated it into our workflow.
> How would you know a non-slop PR didn't use AI?
I don't, hence why I have to take your word for it.
The PRs that people have submitted where they either told me up front or admitted to using AI after the review probed as to why they would be inconsistent in their library usage were not good and required substantial rework.
Yes, some people may submit PRs that used AI and were good. But if so, they haven't told me but I would have hoped that people advocating it would have either told me or got me to review, said it is good, and then told me it was a test and the AI passed. So far that hasn't happened, so I'm not convinced it's a regular occurrence.
Maybe the problem with understanding the benefits of AI is that you are relying on other people to use AI properly. As the direct user myself, I don't have that problem.
I'm using it to make things better rather than just producing. Even just putting it in agent mode and saying "look at all my code and tell me where I can make it better" is an interesting exercise. Some suggestions I take, some I don't.