recallA retrieval
field guide

Follow the question.
Build the understanding.

Every lesson, lab and knowledge check.
Work through in order, or follow your curiosity.

A self-paced, text-first course for learners comfortable with Python. No background in search or machine learning is assumed: each idea is built up from first principles, then tested in a lab that runs in your browser with no API key.

018 lessons

Foundations of Information Retrieval

How search engines find and rank documents, how to measure them, and why LLMs need retrieval.

Explore unit
  1. How Search Systems Evolved
  2. Boolean Search and Inverted Indexes
  3. Ranking with TF-IDF and BM25
  4. Measuring Retrieval Quality
  5. Where Lexical Search Falls Short
  6. Why LLMs Need External Knowledge
  7. RAG, Prompting, Fine-tuning or Long Context
  8. Anatomy of a RAG System
  9. Unit 1 knowledge check
028 lessons

Embeddings and Semantic Retrieval

Turn text into vectors that capture meaning, and choose the right embedding model.

Explore unit
  1. What Embeddings Are
  2. How Embedding Models Learn Meaning
  3. Similarity Metrics
  4. Modern Embedding Models
  5. Choosing an Embedding Model
  6. Multilingual and Indic Retrieval
  7. Sparse vs Dense Retrieval
  8. Building Semantic Search from Scratch
  9. Unit 2 knowledge check
038 lessons

Document Processing

Turn PDFs, slides, scans and web pages into clean, structured text.

Explore unit
  1. Enterprise Documents and Their Formats
  2. Parsing Documents into Markdown
  3. OCR for Scanned Documents
  4. Extracting Tables, Forms and Figures
  5. Cleaning and Normalizing Text
  6. Handling Sensitive Data at Ingestion
  7. Document Connectors
  8. Building an Ingestion Pipeline
  9. Unit 3 knowledge check
048 lessons

Knowledge Base Engineering

Chunk, label and maintain a knowledge base that stays useful as documents change.

Explore unit
  1. Designing a Knowledge Base
  2. Chunk Size and Overlap
  3. Recursive and Structure-aware Chunking
  4. Semantic Chunking
  5. Parent-Child Chunking
  6. Late Chunking
  7. Metadata Engineering
  8. Keeping the Index Fresh
  9. Unit 4 knowledge check
0512 lessons

Vector Databases and Vector Search

Store and search vectors at speed with FAISS and PostgreSQL + pgvector.

Explore unit
  1. Why Vector Databases Exist
  2. Exact vs Approximate Nearest Neighbour Search
  3. ANN Index Types
  4. Vector Search with FAISS
  5. PostgreSQL + pgvector
  6. Schema Design for Vector Storage
  7. Vector Database Operations
  8. Tuning Retrieval Parameters
  9. Hybrid Search and Rank Fusion
  10. Filtered Vector Search
  11. Vector Compression
  12. Choosing a Vector Database
  13. Unit 5 knowledge check
0610 lessons

RAG Engineering

Build RAG pipelines with citations, query rewriting, reranking and agentic retrieval.

Explore unit
  1. Building a Naïve RAG Pipeline
  2. RAG with LangChain
  3. Constructing the Context and Prompt
  4. Citations and Grounded Answers
  5. Query Rewriting, Expansion and HyDE
  6. Multi-query and Self-query Retrieval
  7. Reranking Retrieved Results
  8. Conversational RAG
  9. Query Routing
  10. Agentic RAG
  11. Unit 6 knowledge check
079 lessons

Advanced Retrieval Architectures

Go beyond naïve RAG with contextual retrieval, GraphRAG and structured data.

Explore unit
  1. Where Naïve RAG Breaks
  2. Contextual Retrieval
  3. Knowledge Graphs for Retrieval
  4. GraphRAG
  5. Multi-vector Retrieval
  6. Retrieval over Structured Data
  7. Long Context vs RAG
  8. Emerging Retrieval Paradigms
  9. Choosing a Retrieval Architecture
  10. Unit 7 knowledge check
088 lessons

Multimodal Knowledge Systems

Retrieve from tables, charts, page images, audio and video.

Explore unit
  1. Multimodal Enterprise Knowledge
  2. Vision-Language Document Parsing
  3. Multimodal Embeddings
  4. Cross-modal Retrieval
  5. Page-image Retrieval
  6. Multimodal Knowledge Base Design
  7. Multimodal RAG
  8. Audio and Video Retrieval
  9. Unit 8 knowledge check
0910 lessons

Debugging and Evaluating RAG Systems

Find out why answers go wrong and measure quality with real evaluation sets.

Explore unit
  1. How RAG Systems Fail
  2. Building an Evaluation Dataset
  3. Benchmarking Retrieval Configurations
  4. Evaluating Generated Answers
  5. Evaluation Frameworks
  6. Debugging Chunking and Embeddings
  7. Debugging Retrieval
  8. Debugging Generation
  9. Error Analysis
  10. Regression Testing for RAG
  11. Unit 9 knowledge check
1011 lessons

Production Retrieval Systems

Scale, secure, monitor and control the cost of retrieval in production.

Explore unit
  1. Tracing and Observability
  2. Monitoring in Production
  3. Context Compression and Token Budgeting
  4. Performance Tuning for Vector Search
  5. Scaling to Millions of Documents
  6. Caching for RAG
  7. Cost Engineering
  8. Deployment Architectures
  9. Permission-aware Retrieval
  10. Securing RAG Applications
  11. Production Readiness
  12. Unit 10 knowledge check

Follow your curiosity.

↑ ↓ to explore · Enter to open · Escape to close