Skip to content

1.1 Before You Begin

This is a crash-course on the Azure AI Agent Service with hands-on lab exercises.

AI Agents

1. Learning Objectives

By the end of this crash course, I should be able to:

  1. Describe what the Azure AI Agent Service does
  2. Setup an Azure AI project to use Agent Service
  3. Create, configure, and run, my first AI Agent
  4. Implement responsible AI principles for agents

2. Lab Prerequisites

I checked off these pre-requisites before I got started:

  • An active Azure subscription - with registered resource providers
  • A personal GitHub account - run codespaces on this repo.
  • Azure CLI installed locally - configured in dev container.
  • Familiarity with Python and Visual Studio Code

3. Learning Environment

This repository is set up with a dev container that contains all the necessary tools and library dependencies required for doing the hands-on exercises - including extensions for the Visual Studio Code IDE (my preferred development environment).

To get started with this pick one option below. I recommend GitHub Codespaces. It's easier to get started with, and reproducible across teams for debugging - but requires networking connectivity and codespaces quota.

  1. Fork the repo to your profile.
  2. Open the fork in your browser - click the Code tab.
  3. Create a new Codespaces on main.

This sets up a development container in the cloud with a Visual Studio Code editor in the browser.

  1. Fork the repo to your profile.
  2. Clone the forked repo to your local device.
  3. Install Docker Desktop on device, and launch it.
  4. Install Visual Studio Code on device, and launch it.
  5. Install the Dev Container extension on Visual Studio Code.
  6. Open the cloned repo in Visual Studio Code.
  7. You will be prompted to "Reopen in container" - confirm it.

This launches a Visual Studio Code instance on local device, with your code, connected to a dev container in Docker Desktop.

4. Learning Resources

  1. Azure AI Agent Service Documentation
  2. Azure AI Foundry Documentation
  3. Azure CLI Documentation
  4. Azure resource providers and types
  5. Azure Cognitive Services - code-first modules for Python
  6. Azure AI Samples - for diverse scenarios