Hands-On AI: RAG using LlamaIndex

Contents

• Introduction Introduction

  • Overcome the limitations of LLMs with RAG

    52s
  • Limitations of LLMs

    4m 7s
  • Use cases for retrieval-augmented generation (RAG)

    5m 55s
• 1. Getting Started 1. Getting Started

  • Using GitHub Codespaces

    2m 59s
  • Setting up your environment

    9m 55s
  • Choosing an LLM and embeddings provider

    8m 11s
  • Setting up LLM accounts

    3m 31s
  • Choosing a vector database

    5m 22s
  • Setting up a Qdrant account

    4m 14s
  • Downloading our data

    4m 49s
• 2. Fundamental Concepts in LlamaIndex 2. Fundamental Concepts in LlamaIndex

  • How LlamaIndex is organized

    6m 11s
  • Using LLMs

    8m 48s
  • Loading data

    11m 52s
  • Indexing

    8m 11s
  • Storing and retrieving

    8m 55s
  • Querying

    9m 8s
  • Agents

    6m 14s
• 3. Introduction to RAG 3. Introduction to RAG

  • Components of a RAG system

    6m 25s
  • Ingestion pipeline

    6m 11s
  • Query pipeline

    10m 34s
  • Prompt engineering for RAG

    8m 44s
  • Data preparation for RAG

    18m 38s
  • Putting it all together

    7m 40s
  • Drawbacks of Naive RAG

    4m 47s
• 4. RAG Evaluation 4. RAG Evaluation

  • Introduction to RAG evaluation

    5m 53s
  • Evaluation metrics

    8m 16s
  • How to create an evaluation set

    11m 23s
• 5. Advanced RAG: Pre-Retrieval and Indexing Techniques 5. Advanced RAG: Pre-Retrieval and Indexing Techniques

  • How we can improve on Naive RAG

    4m 24s
  • Optimizing chunk size

    25m 59s
  • Small to big retrieval

    15m 12s
  • Semantic chunking

    11m 45s
  • Metadata extraction

    12m 14s
  • Document summary index

    10m
  • Query transformation

    8m 47s
• 6. Advanced RAG: Post-Retrieval and Other Techniques 6. Advanced RAG: Post-Retrieval and Other Techniques

  • Node post-processing

    15m 35s
  • Re-ranking

    4m 33s
  • FLARE

    8m 44s
  • Prompt compression

    5m 57s
  • Self-correcting

    13m 15s
• 7. Modular RAG 7. Modular RAG

  • Hybrid retrieval

    14m 32s
  • Agentic RAG

    14m 48s
  • Ensemble retrieval

    5m 51s
  • Ensemble query engine

    6m 20s
• Conclusion Conclusion

  • LlamaIndex evaluation

    1m 56s
  • Comparative analysis of retrieval-augmented generation techniques

    8m 8s