This post discusses the challenges of integrating AI agents into the code review process within software development teams, highlighting the principal-agent problem and suggesting that small, high-trust teams can benefit from such integrations, while larger organizations face significant obstacles due to lower trust levels.