Build Production-Ready Agents on Google Cloud
This 1-day course is an advanced, hands-on look at taking AI agents from prototype to production on Google Cloud.
In Build Production-Ready Agents on Google Cloud you build on what you already know about the Agent Development Kit (ADK) and put Google’s agentic stack to work. You start with a review of the stack, the role of the model as the agent’s brain, open source libraries and protocols, and the available deployment targets. Then you learn how to manage context with sessions, memory and the example store, and build an ADK agent that uses all three in a lab.
The course continues with the Model Context Protocol (MCP), covering both using MCP tools with ADK and creating MCP servers, and with the Agent2Agent (A2A) protocol for multi-agent solutions, including a lab where you connect to remote agents using ADK and the A2A SDK. It closes with designing human-agent interactions: how to communicate with agents, control access and build client experiences around them.
This course is aimed at generative AI developers, AI engineers and agentic system architects. Participants should have completed Deploy Multi-Agent Systems with Agent Development Kit and Agent Engine or have equivalent knowledge, along with Python, prompt engineering and ADK experience.