I have never tried seriously coding with AI. I just ask ChatGPT for snippets that I can verify, to save a few round trips to Google and API docs.

However the other day I gave ChatGPT a relatively simple assignment, and it kept ignoring the rules. Every time I corrected it, it broke a different rule. I was asking it for gender-neutral names, but it kept giving last names like Orlov (which becomes Orlova), or first names that are purely masculine.

Is it the same with vibe coding?

Has been my experience as well, never used ChatGPT for anything but as an interface to external documentations and understanding unfamiliar syntax.

Tried using it for the first time for vibe coding and was quite disappointed with the overall result, felt like a college student hastily copy pasting code from different sources for a project due tomorrow.

Maybe I just gave bad prompts…

It's the same. Every assumption, taste, doc, opinion, edge case, unstated knowledge, requirement, exception to the requirement, intention, should be provided for best results. It's not crucial for throwaway weekend projects, but for harder things, it is.

I find it to be the most challenging part. There's a large amount of unstated assumptions that you take for granted, and if you don't provide them all upfront, you'll need to regenerate the code, again and again. I now invest a lot of time into writing all this down before I generate any code.