Agent Operations on Google Cloud
This 1-day course is about taking AI agents from a working prototype to a well-run production service on Google Cloud.
In the Agent Operations on Google Cloud course, you will build on your experience with the Agent Development Kit and learn how to operationalize agents using AgentOps. After a look at the challenges of managing production agents and their deployment targets, the course covers building CI/CD pipelines for agents and using a governed artifact management ecosystem. You will also implement evaluation systems with the GenAI Evaluation Service and ADK, and apply observability for debugging and continuous improvement with Cloud Logging and Cloud Trace.
The day closes with the safety and cost side of running agents: adding guardrails against agent-specific threats with Model Armor and Sensitive Data Protection, and applying FinOps strategies to understand and manage what your agents cost. The course is instructor-led and organized in 6 modules.
This advanced course is aimed at application developers, DevOps engineers, ML engineers and anyone deploying agentic applications on Google Cloud. Attendees should have completed Deploy Multi-Agent Systems with Agent Development Kit and Agent Engine or have equivalent knowledge, and be comfortable with Python, prompt engineering and the Agent Development Kit.