Observability for AI Agents

Why watch

Demonstrates implementing three-layer observability for AI agents using Strands Agents: metrics collection (cycles, tokens, tool timing), distributed tracing (full decision paths with inputs/outputs), and custom trace attributes via hooks. Shows production deployment with AgentCore and CloudWatch GenAI Observability using a travel agent demo that searches flights, checks weather, and books trips. Ideal for developers building production AI agents who need visibility into agent behavior and performance.