Hacker News Reader

I mostly just approve PRs because I trust my engineers. I have developed a 6th sense for thousand-line PRs and knowing which 100-300 lines need careful study.

Yes I have been burned. But 99% of the time, with proper test coverage it is not an issue, and the time (money) savings have been enormous.

"Ship it!" - me

View on HN

I think this points out the crux of the difference of collaborating with other devs vs collaborating with am AI. The article correctly States that the AI will never learn your preferences or idiosyncrasies of the specific projects/company etc because it effectively is amnesic. You cannot trust the AI the same you trust other known collaborators because you don't have a real relationship with it.

Most AI coding tools are working on this problem. E.g. say with Claude Code you can add your preferences to claude.md file. When I notice repeatedly correcting AI's mistake I add instruction to claude.md to avoid it in the future. claude.md is exactly that: memory of your preferences, idiosyncrasies and other project-related info.

I do something to the effect of "Update LLM.md with what you've learned" at the end of every session, coupled with telling it what is wrong when I reject a change. It works. It could work better, but it works.

Haha, doing this with AI will bury you in a very deep hole.

Reply on news.ycombinator.com