loading
loading
Reverting a release, migration, model, or config change to the last known-good state after something goes wrong.
Rollback is the escape hatch for shipping: if error rates spike, quality drops, or a migration corrupts data, you move back to the previous safe version instead of debugging in the blast radius. Good systems make rollback boring with versioned deploys, reversible migrations, feature flags, and model-routing fallbacks. Example: a new agent prompt starts sending bad tool calls, so you switch traffic back to the previous prompt version while investigating.
Plainly
Think of Rollback as the checklist that keeps the app open for real people. Reverting a release, migration, model, or config change to the last known-good state after something goes wrong.
In practice
Use it when a change has to survive deploys, users, incidents, analytics, or billing reality. In practice, define the owner, input, output, and failure mode before you rely on it.