AI agents are moving from toy demos to serious developer tools, and the real power is when you can have them taken action. At Google Cloud, we’ve been building this future through the gcloud MCP, a bridge between AI agents and Cloud APIs. Using MCP, developers can query gcloud using natural language and give agents a structured layer to call.
This talk walks through how MCP servers work. You’ll learn the basics of writing your own MCP server, step through concrete examples, and see how the same patterns apply across different AI agents like Gemini, Claude, and Cursor. Along the way, we’ll cover practical use cases that you can adapt to your own environment.
Finally, we’ll share lessons we’ve learned applying AI to development at scale, and how to plug AI into your engineering workflow.
Speaker
Julie Qiu
Uber Tech Lead, Google Cloud SDK @Google, Building Client Libraries and Command Line Tools Across Different Language Ecosystems
Julie Qiu is the Uber Tech Lead for the Cloud Software Development Kit (SDK) at Google, where she builds client libraries and command line tools across different language ecosystems to interact with Google Cloud products. Previously, Julie was a tech lead on the Go Security team, where she spearheaded Go's support for vulnerability management and Go's package discovery site, pkg.go.dev. She lives in New York City, but loves to spend her time traveling the world.
