Augment your LLM Using Retrieval Augmented Generation

 

About this Course

Retrieval Augmented Generation (RAG) - Introduced by Facebook AI Research in 2020, is an architecture used to optimize the output of an LLM with dynamic, domain specific data without the need of retraining the model. RAG is an end-to-end architecture that combines an information retrieval component with a response generator. In this introduction we provide a starting point using components we at NVIDIA have used internally. This workflow will jumpstart you on your LLM and RAG journey.

What is RAG?

Retrieval Augmented Generation (RAG) is an architecture that fuses two powerful capabilities:

Information retrieval (like a search engine)

Text generation (using an LLM)

Instead of relying solely on a model’s pre-trained knowledge, RAG retrieves external, real-time or domain-specific information and injects it into the prompt. This results in:

• More accurate and up-to-date responses
• Customization to private/internal knowledge bases
• Better transparency and fact-grounding

Learning Objectives

By the end of this course, you will be able to:

Explain the Concept of Retrieval Augmented Generation (RAG):

Understand how RAG enhances LLM outputs by integrating external data sources during inference.

Describe the Components of a RAG Pipeline:

Break down the key stages—retrieval, prompt construction, and generation—and how they interact.

Implement a Simple RAG Workflow:

Build a working prototype that indexes documents, performs semantic search, and feeds relevant context to a language model for generation.

Use Open-Source Tools for RAG:

Get hands-on with libraries such as FAISS, Hugging Face Transformers, and simple vector stores to create a full retrieval-to-generation loop.

Evaluate the Benefits and Limitations of RAG:

Assess use cases where RAG is most effective, and understand its trade-offs (e.g., latency, relevance, hallucination reduction).

Topics Covered

Introduction to RAG

• What is Retrieval Augmented Generation?
• Why use it with LLMs?

RAG Architecture Overview

• Separation of retrieval and generation
• Benefits over pure LLM prompting

Data Indexing and Retrieval

• Creating vector embeddings
• Using FAISS or similar vector stores
• Semantic search vs keyword search

Prompt Augmentation

• Injecting retrieved documents into prompts
• Context window management

LLM Integration

• Feeding augmented prompts into LLMs
• Generating responses with grounded context

Hands-On Lab: Build a RAG Pipeline

• Index a document set
• Perform retrieval
• Generate RAG responses

In the age of LLMs, accuracy, context, and traceability matter more than ever. RAG enables smarter, leaner, and more trustworthy AI—especially in enterprise and mission-critical applications.

With this course, NVIDIA DLI has created one of the most accessible and practical introductions to RAG currently available. It’s short, impactful, and leaves you with working code and a real-world understanding of how to augment your AI with knowledge.

Free Courses : Augment your LLM using RAG