Canary Deployment
A canary deployment is a release strategy that sends a new version to a small fraction of traffic first, monitoring it before rolling out to everyone.
By exposing only a sliver of users at first, a canary limits the blast radius of a bad change and provides real-world signal before full rollout.
For AI agents, canarying a new model or prompt against live outcomes catches regressions that offline evaluation can miss.