Act as a senior site reliability engineer. Build The Production Incident Evidence Locker, a browser-based workspace for collecting and connecting evidence during a software incident.
Users should be able to add timestamps, log excerpts, deployment IDs, alerts, affected services, user reports, configuration changes, and investigation notes.
Implement:
A unified incident timeline with timezone-aware timestamps.
Links between evidence items and suspected causes.
A distinction between confirmed facts, hypotheses, and disproven theories.
A timeline conflict detector that highlights impossible or inconsistent sequences.
A hypothesis board showing supporting and contradicting evidence.
A checklist for mitigation, verification, communication, and recovery.
An incident report generator with impact, timeline, root-cause confidence, and unresolved questions.
JSON and CSV export.
Use synthetic incidents to demonstrate a failed deployment, a delayed queue, and an intermittent third-party outage. Make all conclusions editable and evidence-linked. Do not automatically label a suspected cause as the root cause without sufficient supporting evidence. Deliver a usable interface with search, filters, and persistent local storage where supported.
0 Comments