Step-by-Step AI Agent Tutorial: Build, Test, and Deploy with ADK
Start building your own AI agents with the Startup technical guide: AI agents, learn to build from prototype to deployment. Download guide: https://goo.gle/47rX2gA
You've learned the components, now let's put them to work. In this video, we bridge the gap between theory and a real, working AI Agent with a complete, hands-on demo.
Follow along with Ivan as he guides you through the entire lifecycle of an AI Agent—from local development using the Agent Development Kit (ADK) to deploying it for scalable use on Google Cloud. See how the integration of ADK, MCP, and A2A comes to life in a practical project.
💻 In This Demo, You Will Learn How To:
Build a working agent from the ground up using ADK.
Switch between different Large Language Models (e.g., swapping from Claude to Gemini).
Test & Evaluate your agent's performance by creating an evaluation set to measure actual vs. expected results.
Enable your agent with a real-world tool to pull trending topics directly from Reddit.
Deploy and scale your locally-built agent on Google Cloud for your users.
This is your chance to see how the pieces fit together and get an idea of how to experiment with and build your own agents.
Presented by: Ivan Nardini, Developer Relations Engineer, AI/ML, Google Cloud
▶️ Watch the full webinar: Startups technical guide: AI Agents https://goo.gle/4he2gjw
🧑💻 Explore ADK Code: https://goo.gle/4hHtIpZ
View Sample Codes on GitHub: https://goo.gle/3Jvh9S1
What's the first real-world tool you would give to your AI Agent? Share your ideas in the comments!
👍 Don't forget to like and subscribe for more AI insights!