Microsoft Foundry has become Microsoft’s unified platform for building, deploying, and governing AI applications. It consolidates what used to be a fragmented landscape of tools and services into a single, end‑to‑end developer environment, streamlining the entire AI lifecycle.

From experimentation to production, offering integrated model access, enterprise security, and managed deployments without the usual operational overhead. This shift enables developers to focus on building intelligent features instead of managing infrastructure complexity.

In this talk, I’ll walk you through how Microsoft Foundry works from a developer’s perspective: how to use the Microsoft Foundry SDK to access models through a single interface, combine them with AI services like Speech, Vision, and Document Intelligence. We'll also look at how development and deployment comes together, from building a solution locally to deploying it directly into Foundry's managed infrastructure.

By the end of this session, you’ll understand how to build AI applications faster and more reliably using Microsoft Foundry. How to use its SDK to access models and services through a unified interface, how to structure your projects for real‑world deployment, and how to take advantage of its managed platform to ship production‑ready intelligent features without worrying about servers, scaling, or complex orchestration