Azure-AI_Foundry

Module 1: Introduction to Azure AI Foundry

Overview

Azure AI Foundry (formerly Azure AI Studio) is Microsoft’s unified platform for building, evaluating, and deploying generative AI solutions. It brings together the full lifecycle of AI development — from model selection and prompt engineering to evaluation, deployment, and monitoring — into a single, integrated experience.

📖 Official docs: What is Azure AI Foundry?

What Is Azure AI Foundry?

Azure AI Foundry is a web-based portal (ai.azure.com) and SDK ecosystem that provides:

The Name Change

In November 2024, Microsoft rebranded Azure AI Studio to Azure AI Foundry to better reflect its role as a comprehensive AI development platform. The underlying services and APIs remain the same.

Key Capabilities

1. Model Catalog

Browse and deploy 1,600+ models from leading providers. Compare benchmarks, review model cards, and deploy with a few clicks.

2. Playground

Interactively test models with different prompts, parameters, and configurations before writing any code.

3. Prompt Flow

Build complex AI workflows by chaining LLM calls, Python code, and tools into orchestrated pipelines.

4. Evaluation

Measure model quality with built-in and custom metrics — groundedness, relevance, coherence, fluency, and safety.

5. Deployment

Deploy models as managed endpoints with built-in scaling, monitoring, and content filtering.

6. AI Agents

Build autonomous agents that use tools, access data, and execute multi-step tasks using the Azure AI Agent Service.

Architecture: Core Concepts

┌─────────────────────────────────────────────┐
│                  Azure AI Hub                │
│  (Shared infrastructure & governance)        │
│                                             │
│  ┌──────────────┐  ┌──────────────┐         │
│  │  Project A   │  │  Project B   │         │
│  │  - Models    │  │  - Models    │         │
│  │  - Data      │  │  - Data      │         │
│  │  - Flows     │  │  - Flows     │         │
│  │  - Endpoints │  │  - Endpoints │         │
│  └──────────────┘  └──────────────┘         │
│                                             │
│  Connections: Azure OpenAI, AI Search,      │
│               Storage, Key Vault            │
└─────────────────────────────────────────────┘
Concept Description
Hub Top-level container providing shared infrastructure, security, and connections
Project A workspace within a hub where teams build and deploy AI solutions
Connection A link to an external resource (Azure OpenAI, AI Search, Storage, etc.)
Deployment A model hosted as an API endpoint for consumption

When to Use AI Foundry vs. Standalone Services

Scenario Recommendation
Building a multi-model AI application ✅ AI Foundry — unified management
Simple single-model API call ⚡ Standalone Azure OpenAI Service
Enterprise team collaboration ✅ AI Foundry — RBAC, shared resources
RAG with Azure AI Search ✅ AI Foundry — integrated connections
Custom vision / speech only ⚡ Standalone Azure AI Services
Full AI lifecycle (build → evaluate → deploy → monitor) ✅ AI Foundry

What You’ll Learn in This Module

Lesson Topic
Lesson 1 Hub & Project architecture, underlying resources, networking
Lesson 2 Creating your first hub and project, portal walkthrough, SDK setup
Knowledge Check 15 scenario-based review questions

Prerequisites

Resource URL
AI Foundry Portal https://ai.azure.com
What is AI Foundry? https://learn.microsoft.com/en-us/azure/ai-studio/what-is-ai-studio
Quickstart https://learn.microsoft.com/en-us/azure/ai-studio/quickstarts/get-started-playground
SDK Reference https://learn.microsoft.com/en-us/python/api/overview/azure/ai
Pricing https://azure.microsoft.com/pricing/details/ai-studio/

Next: Lesson 1 — Foundry Architecture →