Chapters
- 01 What Is a RAG Pipeline? A Practitioner's GuideWhat is a RAG pipeline? Why RAG suits private data better than fine-tuning, the ingestion, embedding, retrieval and generation stages, and how it fails.
- 02 LLM Fine-Tuning vs RAG: The 2026 GuideFine-tuning versus RAG for LLMs: what each changes, where each fails, when to combine them, a decision tree, pitfalls and a HyDE retrieval example.
- 03 Vector Database Comparison 2026: What Actually Works in ProductionHow to choose a vector database: extend your existing database or go purpose-built, size the workload, filtering needs, hybrid search and operations.
- 04 How to Evict Stale Embeddings from CacheHow to keep an embedding cache fresh: why TTLs fall short, then versioning, dual-write, source-aware invalidation, drift detection and skipping caching.
- 05 Is ChatGPT a RAG LLM?Is ChatGPT a RAG system? It is an instruction-tuned LLM that sometimes retrieves. This covers the difference, when RAG matters, and when to build your own.