This agent was built using the Google Cloud ADK and Gemini 1.5 Flash as part of the Gen AI Academy APAC 2026.
Google Cloud Gen AI Academy APAC 2026 - Track 1 This project is a production-ready AI agent built using the Google Cloud Agent Development Kit (ADK) and Gemini. It was developed as part of the APAC Gen AI Academy to demonstrate how to move from a local Python prototype to a scalable, serverless microservice.
Project Motive The goal of this project was to "Scale AI Agents across APAC" by using modern cloud infrastructure. Instead of a simple chatbot, this agent uses Tool-Calling logic to provide specific, structured information.
The Tech Stack
Model: Gemini (via Vertex AI)
Framework: Python & Google ADK
Infrastructure: Google Cloud Run
Environment: Google Cloud Shell
Key Learnings & Takeaways
Production Architecture: Learned how to structure a Python project specifically for ADK deployment.
Tool use Logic: Implemented an agent that doesn't just talk but uses tools to fetch data.
How it Works
The user sends a query to the Cloud Run endpoint.
The ADK logic processes the request and decides which tool to use.
Gemini generates a natural language response based on the tool's output.
The serverless container handles the request and then goes back to "sleep" to save resources.
Note: This project was built and verified during the Gen AI Academy APAC 2026 organized by Google Cloud and Hack2Skill.