10 hours ago · 22 min read4339 words · Tech · hide

Most of my repositories have exactly one contributor. That contributor also reviews the pull requests, approves them, and merges them, usually within the same minute. It is a very efficient process. It is not a very good one. GitHub users have had AI review bots for a while now. My code lives on a self-hosted Forgejo instance, and I was not interested in giving a SaaS product a token for it. So I built my own: a Forgejo PR review bot that reads every new pull request, asks an LLM for a review, and posts the result as ordinary comments. The first version took an afternoon. Making it useful took three more rounds, and nearly all of that work went into a single problem: the model was very confident about code it had never seen. Table of Contents Table of Contents What It Does Architecture Line Numbers the Model Can Actually Use Chunking and Cost Guards Reasoning Models and the Empty Answer The Real Problem: Confident Guesses The Verification Pass Precision Is Not the Whole Story Least…

Commenting on Bubbles has been removed. Why? If you want to talk about this post, write to the author, use the comments on their blog if it has them, or reply with a "RE:" post on your own blog. Suggestions? feedback@bubbles.town