The AI landscape is exploding - hundreds of models, dozens of SDKs, and a new framework every week. How do you cut through the noise and actually ship something? In this session, we start from scratch and build a fully functional AI-powered application using Azure AI Foundry. You will see how to browse the Model Catalog, deploy a model in minutes, iterate on prompts in the Playground, and then wire everything into a real app using the Microsoft Foundry SDK. Along the way, we will add grounding data with Retrieval Augmented Generation (RAG), enable streaming responses for a snappy UX, and light up tracing so you can debug your AI calls the same way you debug HTTP requests today. By the end of the session, you will have a repeatable recipe for going from “I have an idea” to “it’s running in production” - with content safety, cost management, and observability baked in from day one.

This session is rated level 100 -"Introductory/Overview".

Reference Links

Interested in this Talk?

Would you like this talk at your event? You can send me an email. If you use Sessionize you can view the talk on Sessionize.

Updated: