All courses

RAG Systems

Build a question-answering app over your own documents, one small lesson at a time: loading, chunking, embeddings, hybrid search, reranking, evaluation, caching, security, access control and tracing. It runs on a basic laptop with NVIDIA's free hosted models. You do not write code. Your teacher writ

What you will learn

  1. Introduction
  2. What Is RAG, and When Not to Use It
  3. Loading Documents
  4. Chunking
  5. Embeddings
  6. Vector Store
  7. Retrieve and Answer
  8. Grounded Answers
  9. Hybrid Search
  10. Reranking
  11. Contextual Chunks
  12. Parent-Child Chunks
  13. Follow-up Questions
  14. Query Expansion
  15. Measure Retrieval
  16. Measure Answers
  17. Caching Layers
  18. Semantic Cache
  19. Fresh Caches and Prompt Caching
  20. Rate Limits, Cost and Speed
  21. Agentic RAG
  22. Facts, Tables and Graphs
  23. Images in RAG
  24. Defending Against Poisoned Documents
  25. Access Control and Leaky Caches
  26. Ingestion Pipelines
  27. Observability and Capstone