What if you could build and orchestrate powerful AI features without managing heavy infrastructure or relying on closed APIs? In this session, we’ll show how to combine the lightweight Gemma model, serverless Cloud Run, and Firebase Genkit to create fast, flexible, and production-ready AI systems.

You’ll learn how to deploy Gemma as a scalable inference service on Cloud Run, integrate it into your app with Genkit flows, and build advanced capabilities like tool-calling, multi-step reasoning, and Retrieval-Augmented Generation (RAG). This setup gives you full control over the AI stack—while staying cost-efficient and developer-friendly.

Ideal for developers building modern apps with AI on their own terms.