Observability
The three signals a service emits, and instrumenting them so the question you didn't anticipate is still answerable.
6 topics
Structured Logging
Logging events as objects with a request ID attached, so production questions are queries instead of grep.
core20 minMetrics & the RED Method
The three numbers that describe any request-driven service, why you measure percentiles rather than averages, and what to alert on.
core20 minTracing & Context Propagation
Following one request across services and awaits — spans, trace context headers, and sampling that keeps the interesting traces.
core25 minHealth Checks & Readiness
Three different questions an orchestrator asks — am I alive, can I take traffic, have I started — and why conflating them causes outages.
core20 minObservability Cost & Cardinality
Why telemetry bills rival compute bills, what high cardinality does to a metrics system, and how to cut cost without going blind.
deep20 minObservability Fundamentals
Instrumenting a system so you can answer questions about it you didn't think to ask before it broke.
core25 minshared