MCP Toolbox for Databases: Connecting AI Agents to Enterprise Data

As generative AI moves beyond chat interfaces into autonomous, production-ready workflows, one major challenge remains: how can AI agents securely and efficiently interact with enterprise databases?

In this session, Stenal P Jolly, Senior Cloud Architect at Google Cloud, will introduce MCP Toolbox for Databases—an open-source, enterprise-grade server built around the Model Context Protocol (MCP).

You’ll learn how MCP Toolbox can provide AI agents, developer tools, and production applications with secure access to structured enterprise data without requiring teams to build and maintain custom integration code for every database or application.

The session will explore how MCP Toolbox supports 30+ database engines, including:

• Google Cloud databases such as AlloyDB, BigQuery, and Spanner

• PostgreSQL

• MySQL

• Oracle

• MongoDB

• and many more.

Through practical examples and demonstrations, Stenal will show how developers can connect AI-powered applications to databases while maintaining security, scalability, and performance.

About the Speaker

Stenal P Jolly is a Senior Cloud Architect at Google Cloudn with nearly a decade of experience in software engineering, cloud architecture, microservices, and distributed systems.

Before joining Google in 2021, he worked at organisations including Zoho Corporation, Cisco, and Synamedia, where he developed cloud-connected and edge computing solutions for major global clients. He also co-founded Semantica during his engineering studies.

This session is ideal for data engineers, software developers, cloud architects, AI/ML practitioners, and anyone interested in building production-ready AI agents that interact with enterprise data.