Introduction to retrieval-augmented generation concepts

  • Module
  • 8 Units
Beginner
AI Engineer
Developer
Solution Architect
Student
Microsoft Foundry

Retrieval-augmented generation (RAG) helps generative AI applications produce responses that are grounded in relevant information. In this module, you'll explore how a RAG solution prepares, retrieves, and uses data to generate answers.

Learning objectives

By the end of this module, you'll be able to:

  • Explain the purpose and benefits of retrieval-augmented generation (RAG).
  • Describe how data is prepared for retrieval.
  • Describe how a RAG solution retrieves relevant information and uses it to generate a grounded response.
  • Identify considerations for evaluating and improving a RAG solution.

Prerequisites

Before starting this module, you should have:

  • Familiarity with generative AI concepts and terminology.
  • A basic understanding of large language models (LLMs) and prompts.
Start