WORKSHOP: Agentic AI: Observability in LLM Applications

Extending from the foundations covered in the talk "Building, Deploying, and Managing AI Agents at Scale on Google Cloud", this 2-hour workshop provides a hands-on, in-depth experience into one of the most vital pillars of agentic AI at scale: observability and evaluation. Participants will learn how to track agent behaviors, monitor LLM workflows, and design effective evaluation strategies to ensure reliability. The session combines practical demos with interactive discussions to make complex systems transparent and measurable.