I've built an agent that builds documentation for code bases and you are 100% right. Having big picture documentation is important, but having bite size explanations about why some components work a certain way is more important. Making sure the AI doesn't have to infer behavior from code is really powerful. Even going as low level of reference docs. Even though devs would prefer that a method be self-Explanatory, it helps to also have plain english explanation about what's happening in a class or method.

I'm building a way to automatically keep code base documentation updated. https://github.com/apps/askmanu

Right now we're focusing on reference docs and soon the app will be able to write full documentation content.

We want to focus on incremental changes to docs (one PR at a time) so the content is easy to verify and merge.

This is brilliant. Nice work

I've seen that happen but usually with code bases that are either not very well documented (reference docs) or that have a lot of abstractions and are complicated

I would recommend you make it open source. I've also done this the past as well. Even if it's a project that is not maintained you don't know how it might help someone else.

I checked your website I it's an interesting idea. I think you need some additional copy though, because just landing on your homepage you don't really understand what it does. More exactly I don't know what to use it for. Your comment and GP's coming together gave me a better idea.

What's your target audience? developers?

Fantastic feedback, my gosh, thank you. Yes the home content doesn’t do a good job of explaining what it is. Honestly that’s because a lot of the ideas I typed here are pretty fresh, and my ‘marketplace’ is only implemented on prod as a prototype. I do see the the mismatch in copy and the real utility.

I’ll work on positioning and update the copy to keep in sync with my vision …

And yeah, my target audience is developers, specifically those with not a lot of time but with a lot of ideas.

Gosh thanks again, that’s super excellent feedback.

Not everything in life is about money

You usually fine tune when you want to add capabilities (an output style, json output, function calling, etc). You use RAG to add knowledge

Yeah, there's an open issue in GitHub for it. Hope they fix it soon

Take a look at dspy.ai. Love the project but they need some help with the docs.