Day 5: 2. AI Agent That Queries BigQuery — Google Trends Analyst (ADK + MCP)

This tutorial demonstrates building an AI data analyst agent using the Agent Development Kit (ADK) and Model Context Protocol (MCP) that queries live BigQuery data instead of relying on training data. The agent integrates with Google Trends data to answer real-time analytical questions. The implementation showcases how to connect AI agents to actual databases and external data sources for dynamic, fact-based responses.

Key Points

• •Use ADK (Agent Development Kit) to scaffold and manage AI agent development workflows
• •Implement MCP (Model Context Protocol) to enable agent-to-database communication
• •Connect BigQuery as the data source for live, real-time information retrieval
• •Build query logic that translates natural language questions into SQL for BigQuery
• •Integrate Google Trends data to provide current, trending analytical insights
• •Distinguish between training-data-based responses and live-data-driven agent responses
• •Structure agent prompts to guide data retrieval and analysis workflows
• •Test agent responses against real BigQuery datasets to validate accuracy
• •Use MCP tools/resources to abstract database complexity from the agent logic

Workflow Diagram

Concepts