Join Datadog for an interactive, hands-on workshop where you’ll learn how to evaluate, monitor, and optimize AI agent performance using Agent Observability, Evaluations, and Experiments. Whether you’re developing AI agents or managing production AI applications, you’ll gain practical experience improving the quality, reliability, and effectiveness of agentic workflows.

Throughout the workshop, you’ll configure LLM-as-a-judge evaluations, monitor agent behavior, and identify opportunities to improve performance using real-time observability. You’ll then build and run offline experiments to compare models, analyze tradeoffs, and make data-driven deployment decisions before validating improvements with live evaluation scores.