Every time I see someone using LLMs to code, I want to shake them and ask "If LLMs are supposed to be so good, why isn't your code good?" 馃
Post
Replies:
1
I will say I have experienced one exception to this, which is the very responsive maintainership of Optique, which has successfully introduced feature after feature without any regressions and while maintaining good docs: https://optique.dev/
But it stands out because it seems to be such an exception to other projects that use AI. It's gotten to the point where I will sometimes dismiss a project out of hand if I see Claude has contributed.
