Advanced RAG with LlamaIndex - Metadata Extraction [2025]

Learn advanced Retrieval-Augmented Generation (RAG), focusing on techniques such as metadata augmentation and metadata filtering using LlamaIndex. This will help you improve your RAG models for better performance and accuracy. ## Links --- - Notebook: https://colab.research.google.com/gist/alejandro-ao/75d18a00f0fbe6bacb57083c210178a1/llamaindex-vector-stores.ipynb - AI Engineering Bootcamp (join the waitlist): http://aibootcamp.dev/ - Buy me a coffee (or a beer): https://buymeacoffee.com/alejandro.ao ## What You'll Learn --- - Advanced RAG Techniques: Understand the intricacies of Retrieval-Augmented Generation and how it can boost your AI models. - Metadata Augmentation: Learn how to enrich your data with metadata to improve retrieval and generation processes. Also, control what metadata is sent to your LLM and embedding model. - Hands-On Notebook: Follow along as we explore a detailed Jupyter notebook, showcasing practical applications of these techniques. - LlamaIndex Integration: See how LlamaIndex can be integrated into your workflow to streamline RAG processes. ## Timestamps --- 0:00 Intro 2:21 Quick Explanation of RAG 6:21 Setup 7:07 Extract text from File 12:26 Metadata Selection 22:40 Metadata Extraction 35:46 Create An Index 43:55 Use a Vector Store 50:15 Conclusion