Tracing and Monitoring Your AI Agents

Building an AI agent is only the first step—operating it reliably in production is where the real challenge begins. As AI applications become more complex, developers need better visibility into how their agents perform, why they fail, and how to optimize them for real users.

In this session, you'll learn how to trace AI agent executions, monitor latency and token costs, evaluate response quality, and debug production issues using Comet Opik, an open-source observability framework for Google ADK (Agent Development Kit).

Whether you're building with Gemini, Google ADK, LLMs, or modern AI Agents, this session will introduce practical techniques and best practices for building reliable, observable, and production-ready AI systems. If you're a software engineer, AI developer, or GenAI enthusiast looking to move beyond demos and deploy AI with confidence, this session is for you.

Topics you'll explore:

• AI Agent Observability

• Google ADK & Comet Opik Integration

• Execution Tracing & Debugging

• Latency & Cost Monitoring

• Response Evaluation

• Production Best Practices for AI Agents

Join us to learn how to build AI agents that are not only intelligent—but also reliable, scalable, and ready for production.