Fair, but it would be interesting to see how people are implementing this “write the docs you need to do a better job” logic and putting it into use. I’m playing with this but would love to see someone’s success story. “I did X and now the code is better/its more token efficient/reviewers understand the changes/whatever.”
I just let the LLM write the docs it will read and I don't pay attention to them very much unless I need to debug a problem that it can't solve on its own. I just tell it what areas to focus on, it writes stuff that gets checked in but not really read by humans, it updates the docs when it things change before it changes the code, but can also review all the design stuff to when making code changes.
Sometimes I run into a problem that the LLM can't really handle yet, but I just break the problem up into more docs, tests, and code. So...that usually works, but I admit I'm move more slowly on those problems, and I'm not asking the LLM how to break the problem up yet (although I think we will get there soon).