Design, implement, or review production observability using service-level objectives, metrics, logs, traces, profiles, correlation, dashboards, alerts, and diagnostic context. Apply when system behavior or reliability must be understood from telemetry; minimize sensitive data and cardinality.
