[yaseer]

I've found writing docs and updating docs a great AI use-case.

In my experience documentation generation has a lower error rate than code generation, and the costs of errors are lower too.

I'm not really a big fan of AI agents writing features end-to-end, but I can definitely see them updating documentation alongside pull requests.

[throwawayffffas]

While I agree to an extent, I think it's not ideal. The point of documentation in my opinion is to explain intent. If want to figure out the functionality of something the code is just as good as documentation, arguably better.

AI ,because by default only sees the code, in general describes the functionality not the intent behind the code.

[9rx]

> The point of documentation in my opinion is to explain intent.

Of course, that's what your tests are for: To document your intent, while providing a mechanism by which to warn future developers if your intent is ever violated as the codebase changes. So the information is there. It's just a question of which language you want to read it in.

"Updating docs" seems pointless, though. LLMs can translate in realtime, and presumably LLMs will get better at it with time, so caching the results of older models is not particularly desirable.

[chasd00]

This is one area where i think a LLM can really help. It's not going to produce perfect documentation but it's so much more productive to edit/update docs than create docs from scratch. Staring at a blank screen and getting started on docs is the hardest part in my experience.