
Evolution of AI Code Review

The Future of Code Review: Stop Reviewing Line-by-Line, Start Governing AI Agents - Itamar Friedman
Topic 1 of 11
Published May 4, 2026 · Clip 3:04–7:22 (4:18)
Itamar explains how to deal with the noise of AI tooling by holding two buckets: a vision of the future and low-noise information channels. He then frames code review as essential but evolving, drawing a parallel to code generation's progression from autocomplete (gen 1) to agents and agent teams (gen 3.5). Code review similarly evolves from snippet recommendations to full PR review to multi-repo, multi-agent review where designated agents verify UX against Figma, intent against Jira, architecture docs, and database safety, surfacing only what needs human attention.