Observability
tracing · LLM observability
Tracing every request: input, retrieved context, tool calls, output, cost and latency.
01In short
Tracing every request: input, retrieved context, tool calls, output, cost and latency.
In productionWithout it, debugging an agent is guesswork.
02Video
slot · videoObservability in 90 secondsadd: GUIDES['observability'].video
03Guide
slot · guide
A step-by-step guide for “Observability” goes here. Suggested outline:
- What it is — in one paragraph
- Why it matters in production
- How to do it — 3 to 7 steps
- Pitfalls we see in the field
04Checklist
slot · checklist
Four to eight things a team can tick before go-live.
05FAQ
slot · FAQ
The three questions clients actually ask about “Observability”.
06Related terms
SecurityAudit trailA tamper-evident record of what the system did, with which inputs, on whose behalf.EvaluationDriftPerformance degrading over time because data, users or the model change.OperationsLLMOpsThe practices and tooling to run models in production: versioning, deployment, monitoring, evals, rollback. DevOps for systems that change behaviour.
Where we help · Secure