Building RAG Solutions with Azure AI Foundry (Formerly Azure AI Studio)

Course details

This in-depth course is crafted to empower you with the proficiency to effectively leverage Azure AI Studio for the creation of retrieval-augmented generation (RAG) solutions. The course begins with a systematic walkthrough of the process involved in creating a RAG solution and the fundamentals and principles of RAG, diving into the prerequisites for crafting a RAG solution in Azure AI Studio. This hands-on course with instructor Ziggy Zulueta offers demonstrations on importing and vectorizing data, and integrating this data into a deployed model, showing the process of customizing the default solution using the prompt flow feature. The course wraps up with a guide on testing and evaluating the solution, along with a practical session on testing the REST API endpoints in Postman and using the endpoint in Microsoft Teams through Copilot Studio.

Concepts

0. Introduction

01 - Create a RAG solution with little coding

1. Fundamentals of RAG

02 - The basics of RAG - Adding custom data to your LLM
03 - Understanding tokens - A key factor of costs in your system
04 - Vector embeddings - How words connect to each other
05 - How RAG works - Understanding the process under the hood
06 - RAG high-level architecture - The required components

2. Introduction to Azure AI Foundry

07 - Azure AI Foundry overview - Deploy at scale in a safe, secure, and responsible way
08 - Navigating the Azure AI Foundry
09 - Creating a project in Azure AI Foundry

3. Setting Up Azure AI Foundry

10 - Understanding content filters
11 - Creating content filters
12 - Creating model deployments
13 - Navigating the Playground
14 - Using the Playground and its settings

4. Creating an Index for RAG Using Azure

15 - Creating an index using Azure AI Foundry
16 - Creating an index using Azure AI Search
17 - Understanding retrieval and relevance in Azure AI Search
18 - Testing your index in the Playground

5. Introduction to Azure Prompt Flow

19 - Understanding prompt flow
20 - Create a sample prompt flow for RAG
21 - Evaluation and monitoring metrics
22 - Perform evaluations on your RAG system

6. Deploying a RAG Solution

23 - Deploying the RAG solution using prompt flow
24 - Testing the REST endpoint using Postman
25 - Deploying the REST endpoint to Copilot Studio and Microsoft Teams

Conclusion

26 - Key takeaways
27 - Additional learning

Building RAG Solutions with Azure AI Foundry (Formerly Azure AI Studio)

Authors

Ziggy Zulueta