miniblog.

LLMs are surprisingly good at reducing crash samples. I've had success with "this project crashes my static analysis tool with the following command, try to shrink the repro whilst preserving the crash. Commit each smaller version to a branch".

... and my third refactoring with Cursor changed some function calls that I didn't want it to modify. Subtle. I've had the most success with AI coding tools when I know exactly what I want the output to look like.

Perhaps the biggest driver of docs success is prestige. How cool are the docs tools, and how proud are you to have contributed to the docs? Which tools do this best? Racket docs look pretty great, for example.