Lesson 4 · 7 min
Observability
Logging traces, costs, and quality signals.
Log every LLM call with: model, input/output tokens, latency, tool calls, cache hits, and a task ID. Aggregate to dashboards for cost per task and tail latency. Track quality with sampled human review of outputs.
Quick check
Context & ReliabilitySelect one
Why include a request_id with tool calls that may be retried?