Build multi-agent AI A2A + Cloud Run | Hands On AI (Part 2)

Опубликовано 22 марта 2026, 15:43

GCP credit → goo.gle/handson-ep1-lab2
Codelab & source code → goo.gle/summoner
Try Google ADK → goo.gle/4bWK6la
Watch the first video here. → goo.gle/handson-ai-ep1

Build and deploy a production ready multi-agent AI system using Agent-to-Agent (A2A) Protocol, memory & state management, and Google Agent Development Kit (ADK) on Google Cloud Run. This is part 2 of our hands on RPG themed multi-agent lab series.

🗡️ What developers can learn in part 2:
✅ How to implement Agent-to-Agent (A2A) Protocol for direct agent communication.
✅ How to manage agent memory and state across multi-agent workflows .
✅ How to use agent callbacks to control and customize agent behavior.
✅ How to deploy a multi-agent system to Google Cloud Run.
✅ Production ready patterns for agentic AI on Google Cloud.

More resources:
A2A documentation → goo.gle/4bBSPYN
ADK Sample → goo.gle/4rQKWVn
Cloud Run Documentation → goo.gle/3PxbMnW

🔔 Subscribe to Google Cloud Tech → goo.gle/GoogleCloudTech

Speaker: Ayo Adedeji, Annie Wang
Products Mentioned: Agent Development Kit, Model Context Protocol, Agent-to-Agent Protocol, Cloud Run