Azure AI with Foundry — Course Introduction

“Welcome to the Azure AI Engineering course. Over the next 20 episodes we are going to go deep on every dimension of building AI applications on Azure — from your first Azure OpenAI call to enterprise RAG applications, Copilot Studio chatbots, fine-tuned custom models, and responsible AI governance. This is not a survey course. By the end, you'll have the skills and portfolio to work as an Azure AI engineer and be prepared for the AI-102 certification.”

“Here's the most important context for this course. Before 2025, Azure had separate portals for each AI service — Azure AI Foundry for OpenAI, Speech Studio for Speech, Language Studio for NLP, Vision Studio for Computer Vision. It was fragmented and confusing. Microsoft unified all of these into a single platform called Azure AI Foundry, available at ai.azure.com. Every episode in this course will start from AI Foundry. It is your single entry point to everything Azure AI.”

“Let me orient you inside AI Foundry. When you log in at ai.azure.com, you create a Hub — the shared infrastructure that holds your Azure OpenAI connections, Azure AI Search connections, compute, and security settings. Inside the Hub you create Projects — one per AI application you're building. Within a project you have the Model Catalog for choosing and deploying models, AI Services for pre-built capabilities, Prompt Flow for building and testing AI chains, and Evaluation for measuring quality before shipping to production.”

“This course is the companion to the Azure Cloud Engineering course. They're separate disciplines. The Cloud course covers infrastructure — VMs, networking, containers, DevOps. This course covers AI applications — building, evaluating, and deploying intelligent systems. But they connect: your AI application deployed from Foundry runs on Azure App Service or AKS. Your RAG pipeline stores data in Azure Storage and Azure AI Search. Understanding both makes you a complete Azure engineer. One important distinction: Azure Machine Learning at ml.azure.com is a separate portal from AI Foundry — it's for custom model training from scratch, which we cover in episode 4 of this course.”

“Here's your map for the next 20 episodes. We start with the two foundational services — Azure OpenAI for the language model and Azure AI Search for retrieval — because everything else builds on top of them. Then we go through each Azure AI service in depth. We build a complete RAG application as the first capstone project. Then data engineering for AI — how you get clean data to your models. The second half covers AI Foundry deeply, Copilot Studio for no-code AI apps, fine-tuning for custom models, and responsible AI. We finish with the AI-102 exam prep and career roadmap.”

“What do you need before starting this course? An Azure account — the free tier gives you $200 in credits which covers most of the demos. Basic Python programming knowledge is helpful since we'll write code in most episodes. AI or machine learning experience is not required — we'll explain every concept from first principles. Install VS Code, Python 3.10 or later, and the Azure CLI. For the OpenAI and AI Search demos, you'll spend a small amount — about $10-20 total if you follow along with the demos. Azure AI Foundry itself has a free tier for exploration.”

“What does an Azure AI engineer actually do, and what makes a great one? The best AI engineers bridge the gap between AI capabilities and business value. They know the full AI services portfolio — not just how to call an API but when to use which approach. They can design a RAG system that grounds AI answers in company knowledge. They think about safety and responsible AI from the start, not as an afterthought. They can deploy a model to production, monitor it, and know when to retrain. And critically, they can explain what the AI does to stakeholders in business terms.”

“Let's take a live tour of Azure AI Foundry so you're oriented before we go deep in episode 2. I'll open ai.azure.com, create a Hub and Project, and walk you through every section — the Model Catalog where you'll pick models in every upcoming episode, the AI Services section where Speech, Language, Vision, and Document Intelligence all live now, and Prompt Flow where we'll build RAG applications. Bookmark this URL. This is home base for the entire Azure AI Engineering course.”

“You're oriented. You know what AI Foundry is and why it matters. You have your Azure account set up. And you have the full course roadmap in front of you. In the next episode we go straight into Azure OpenAI Service — deploying GPT-4o, engineering prompts, implementing RAG, and building your first AI-powered application. This is where the real building starts. I'll see you in episode 2.”