Fronting the Azure AI Stack with Azure API Management

Agenda:
6:00 PM - 7:30 PM - Fronting the Azure AI Stack with Azure API Management

Abstract:
Calling Azure AI services directly from your app works — until you're juggling keys across OpenAI, Document Intelligence, and Content Safety, need to throttle usage, track cost per team, or swap a backend without redeploying clients. Azure API Management (APIM) sits in front of the whole Azure AI stack and gives you one managed, secured, observable entry point for all of it.

In this online session we'll walk through standing up an APIM gateway and putting multiple Azure AI services behind it — Microsoft Foundry / Azure OpenAI for models, Document Intelligence for extraction, Content Safety for moderation — and the same pattern extends to any other Azure AI service. Through a live demo you'll see how requests actually flow through APIM (operation matching first, backend policy second) and why that one distinction explains most of the 404s and 401s people hit. We'll cover the configuration that makes it work: importing each API, mapping request paths with `rewrite-uri`, the `api-key` vs `Authorization` header gotcha, enabling CORS for browser clients, and applying rate limits and usage metrics per service.

You'll leave knowing how to front any Azure AI service through a single gateway — consistent auth, governance, and monitoring, without changing client code.

Who should come: Developers, architects, and the AI-curious alike. Some REST API familiarity helps, but come as you are — we start from the ground up, no APIM experience required.