BCS - Building Secure AI Agents: An Enterprise Practitioner's View - Open Source SG

From “Hello World” to enterprise-grade: learn how to build and secure AI agents in production.

In this session, Kishor Patil, Senior Engineering Manager at Best Buy and IEEE Senior Member, shares what he has learned from taking AI agents from early prototypes all the way through to secure, governed enterprise deployments.

This is not a theoretical talk. It is a practitioner’s honest account of the gap between “Hello World” and production-ready, what works, what breaks, and what keeps you up at night when AI agents start making decisions autonomously at scale.

We will cover how to think about Agent Identity and access from day one, how to secure the Model Context Protocol configurations that agents rely on, how to detect problems at runtime before they escalate, and how to build governance that actually holds up under real enterprise conditions.

Whether you are just starting to explore AI agents or already running them in production, you will leave with practical ideas you can use immediately.