Field notes on AI code review worth reading
Opinionated engineering notes on the hard part of AI review: reducing noise, keeping humans in control, and turning model output into one useful GitHub review. For shipped changes, see the changelog.
Recent posts
Deduping reviewer findings without losing signal
How we sort by severity, dedupe by (file, line, lowercased title), and only post inline comments on lines that map to a unified-diff position. Plus: what we threw out and why.
A confidence floor is the cheapest noise filter you have
PR Quorum defaults to a 0.75 confidence floor because the fastest way to earn trust is not posting weak findings in the first place.
AI review should be advisory by default
Why every PR Quorum review is posted with event:COMMENT — never request_changes — and how the advisory framing changes how teams actually use the panel.
Want the system behind these posts?
Install PR Quorum on a repo and see the specialist reviewer panel on your next pull request.