Build a RAG-based Chat App on Azure AI Foundry

This repository contains the source and instructions guide for a workshop on building a RAG-based chat app using Azure AI Foundry. The workshop is derived from the official 3-part tutorial and adapted to provide additional resources and suggestions for self-guided learners.

Pre-Requisites

To get the most from this lab, you will need the following:

1. An Azure subscription - Get one for free
2. A GitHub account - Get one for free
3. Familiarity with VS Code, Github & Azure.
4. Familiarity with Python and Jupyter Notebooks.

Check that your Azure subscription has sufficient quota to deploy the following Azure OpenAI Models:

1. Chat: gpt-4o-mini
2. Embeddings: text-embedding-ada-002

Check that your Azure account has the necessary permissions to make role assignments for relevant Azure AI resources. (e.g., may require a Privileged role like Owner, User Access Admin or RBAC Admin)

Getting Started

This repository is instrumented with a devcontainer.json to give you a development environment with all required dependencies pre-installed. To get started:

1. Fork the repository to your personal profile.

2. Launch Codespaces on that fork. Setup may take a few minutes. Once ready, you will see a Visual Studio Code editor.

3. Wait for the terminal prompt then launch the workshop guide with this command:

   mkdocs serve

   You will be prompted to view this in a browser or within VS Code. Select the browser option to get the workshop guide in its own tab.

4. Open a new terminal window in Visual Studio Code, and use this window for executing further commands from the guide.

Troubleshooting

This section will capture any known issues in using this repo or workshop guide.