SYSTEM 05/14 · THE FIELD GUIDE
Retrieval-Augmented Generation (RAG)
Grounding model answers in your own data — chunking, retrieval, reranking, GraphRAG, and evaluation.
RAG Fundamentals
What RAG is, when to choose it, and your first working pipeline.
Retrieval-Augmented GenerationBEGINNERIngest, Retrieve, GenerateBEGINNERRAG vs Fine-TuningBEGINNERBuild Your First RAG AppINTERMEDIATERAG vs Long ContextINTERMEDIATERAG LimitationsINTERMEDIATERAG Still HallucinatesINTERMEDIATERAG vs Search EngineBEGINNERRAG Cost and LatencyINTERMEDIATERAGFlowINTERMEDIATEtxtaiINTERMEDIATE
Chunking & Ingestion
Where most RAG quality is won or lost: how documents become chunks.
Chunking in RAGBEGINNERStrategies ComparedINTERMEDIATEChunk Size & OverlapBEGINNERCode, Tables & MarkdownINTERMEDIATEParsing PDFsINTERMEDIATEContextual RetrievalINTERMEDIATECleaning Data for RAGBEGINNERSemantic ChunkingINTERMEDIATEUnstructuredINTERMEDIATEDoclingINTERMEDIATELlamaParseINTERMEDIATEChonkieINTERMEDIATE
Retrieval & Reranking
BM25 meets embeddings: hybrid search, RRF, rerankers, and query rewriting.
RetrieversBEGINNERRerankingINTERMEDIATEBM25BEGINNERReciprocal Rank Fusion (RRF)INTERMEDIATEHybrid SearchINTERMEDIATECross-Encoders vs Bi-EncodersADVANCEDMetadata Filtering in RAGINTERMEDIATEQuery RewritingINTERMEDIATEHyDEINTERMEDIATECohere RerankINTERMEDIATEBGE RerankerINTERMEDIATEColBERTADVANCEDSPLADEADVANCED
Advanced RAG Architectures
Agentic RAG, GraphRAG, self-correcting pipelines, and retrieval beyond plain text.
RAG Evaluation
Proving your pipeline works — and debugging it when it doesn't.