Microsoft Reactor

AI agents are evolving from prototypes into enterprise systems—but scaling them requires a new development model. In this session, engineers from Microsoft, Capgemini, and NVIDIA introduce a scalable, "factory" approach to building at scale.

We’ll discuss how to standardize development patterns, integrate GitHub Copilot into engineering workflows, and design reusable multi-agent architectures. You’ll also learn how to operationalize agents with governance, security guardrails, and observability—plus how to upskill teams for this new model. Walk away with practical patterns to move from individual agents to scalable, controlled, production-ready systems.