Master Retrieval-Augmented Generation (RAG) from the ground up: learn how document chunking, vector embeddings, metadata filters, hybrid search, and citation runtimes ground language models in live, verifiable facts.
- 1 RAG in One Picture: What Retrieval-Augmented Generation Really Does Understand why static language models hallucinate outdated facts, and discover how Retrieval-Augmented Generation (RAG) decouples neural reasoning from document storage: injecting live, verifiable company records into the prompt window before answer generation begins. Scheduled · November 24, 2026
- 2 Document Chunking for RAG: Too Big, Too Small, and Just Usable Discover why document chunking is the single most critical lever in your RAG pipeline. Learn how fixed-character slicing breaks tables, explore recursive and semantic boundary splitting, and implement parent-child indexing for high-precision retrieval without context loss. Scheduled · November 25, 2026
- 3 Embedding Models for RAG: Picking Dense, Sparse, and Hybrid Vectors Without Religion Cut through vendor hype to choose the right embedding model for your RAG architecture. Explore why pure dense vectors fail on alphanumeric codes, compare top open and commercial models on MTEB, and implement Reciprocal Rank Fusion to combine dense semantics with sparse lexical precision. Scheduled · November 26, 2026
- 4 Metadata Filtering in RAG: The Secret to Security, Speed, and Zero Stale Data Learn why unconstrained vector search leaks private data and retrieves obsolete company policies. Discover the fatal flaws of post-retrieval filtering, explore the four core metadata dimensions, and implement pre-retrieval index partitioning to enforce strict multi-tenant isolation and role-based access control. Scheduled · November 27, 2026
- 5 Citations and Show Me the Source UX: Building Verifiable RAG Applications Discover why language models fabricate believable citations and how to stop them. Learn the four levels of source attribution granularity, build a verbatim span verification engine in Python, and design interactive source drawer interfaces that earn enterprise user trust. Scheduled · November 28, 2026
- 6 RAG Failure Modes: Stale Docs, Wrong Neighbors, and Confident Nonsense Explore why production RAG systems fail when cosine similarity masquerades as factual truth. Learn the mechanics behind stale document collisions, wrong neighbor vector drift, and inverted hallucinations, and build an automated retrieval triage engine in Python to intercept errors before users see them. Scheduled · November 29, 2026
- 7 Light RAG Evals Non-Engineers Can Run: From Guesswork to Golden Sets Escape subjective vibe checks and take control of RAG quality. Learn how non-technical domain experts can build golden evaluation sets, calculate the core RAG Triad metrics, and run automated evaluation scorecards without touching complex machine learning codebases. Scheduled · November 30, 2026
- 8 When Fine-Tuning or a Bigger Context Window Is Better Than RAG RAG is a powerful tool, but it is not always the right tool. Explore the architectural trade-offs between dynamic retrieval, domain fine-tuning, and massive context windows. Learn the unit economics of token costs versus GPU training, and build an automated architectural selection calculator in Python. Scheduled · December 1, 2026