Skip to content
Observability for AI Agents

Table of contents

Opening

Part 1 — Foundations

The production framing, the telemetry structure, and the signals worth tracking first.

Part 2 — Safety, Privacy & Evaluation

What not to capture, how PII leaks into telemetry, and how evaluation becomes an observability signal.

Part 3 — Security & Cost Controls

Guardrail telemetry, security signals, cost attribution, budget limits, and loop controls.

Part 4 — Hands-on Implementation

LangGraph, OpenTelemetry, Langfuse, LLM spans, tool spans, retrieval spans, and runtime hardening.

Part 5 — Langfuse Workflow

Sessions, users, prompt management, Playground, scores, evaluators, annotation, datasets, and experiments.

Part 6 — Operations

Telemetry tests, pipeline resilience, dashboards, alerts, release gates, and runbooks for operating AI agents day to day.

Part 7 — Advanced Workflows

Subgraphs, subagents, handoff traces, streaming graph updates, and production-safe live workflow visibility.

Reference

This book is licensed under Creative Commons Attribution 4.0 International.