Automate project intake with multi-agent AI using MCP, Google ADK, Cloud Run, and BigQuery
Check out the repo here. → https://goo.gle/4wYiZyE
Learn more about building multi-agent systems here. → https://goo.gle/4dTqNZQ
Join GEAR → https://goo.gle/4wWE3FP
Manually vetting hundreds of project requests is a thing of the past. Imagine receiving every proposal with a built in risk score, resource check, and "Go/No-Go" recommendation—delivered in seconds.
Join Kevin Blanco as he demonstrates how to build a powerful multi-agent system that seamlessly integrates Google Cloud and Asana. Watch along and see a real world example of automating an infrastructure request, returning instant historical pattern analysis and live workspace context without any manual steps.
What you'll learn:
- The architecture: How to securely connect Asana webhooks to Google Cloud Run services.
- Orchestration: Using the Google Agent Development Kit (ADK) to build an orchestrator LlmAgent.
- Parallel execution: Dispatching specialist agents in parallel using the Agent-to-Agent (A2A) protocol.
- Open standards: Leveraging the Model Context Protocol (MCP) to connect Vertex AI agents directly to BigQuery and live Asana work graphs.
- Scalability: Deploying cost effective, pay-per-request agents using individual Cloud Run containers.
- Get started: Ready to see human and AI collaboration in action? The full source code is available on GitHub. Clone the repo, set it up in a few commands, and have this multi-agent system running in your own Google Cloud project today!
Go from prototyping to deploying secure, enterprise grade agents with hands on training and guidance from Google experts with GEAR. → https://goo.gle/4wWE3FP
Watch more How I built my AI agent → https://goo.gle/agent-development
🔔 Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech
#GoogleCloud #Gemini
Speakers: Kevin Blanco
Products Mentioned: Agent Development Kit, Model Context Protocol, Agent2Agent Protocol, Application Default Credentials, Cloud Run, BigQuery, Gemini