debugging-and-error-recovery

To find the root cause of a production incident or build break, follow structured triage steps to reproduce the failure, isolate failing components, and separate symptoms from the actual root cause.

Debugging flaky test failures requires preserving evidence like logs and repro steps, then guiding engineers through root-cause verification to ensure robust fixes and prevent recurrence across environments.

Preserve evidence when diagnosing a runtime incident by collecting logs, repro steps, and screenshots using structured checklists, enabling faster fixes and preventing regression.

Structured triage works for software outages in production environments by enforcing stop-the-line rules and verification steps, applying to test failures, build breaks, and runtime incidents.

Verify a software fix to prevent incident recurrence by applying structured diagnosis and verification steps, re-running validations across tests, builds, and production incidents to guard against regression.