Atlas

Prompt & Context Engineering

Retrieval Augmented Generation

Give the model the documents instead of hoping it memorised them.

You can jump straight in, but this star assumes Context Engineering. Not recommended as a first stop.

Four layers of depth

Each layer ends with a quiz. Finish layer 4 and you own this concept.

  1. L1IntuitionWhy a model that has read the whole internet still can't answer questions about your company, and how retrieval fixes it.6m
    40 XP
  2. L2MechanicsThe pipeline in detail: chunking, embeddings, vector search, hybrid retrieval, reranking, and prompt assembly.9m
    70 XP
  3. L3CodeBuild the whole loop: chunk with overlap, embed, store in pgvector, hybrid search with RRF, and assemble a cited prompt.12m
    110 XP
  4. L4FoundationsThe geometry of similarity, why ANN search is approximate, HNSW's structure, and the recall-latency-memory triangle.12m
    180 XP

33 stars in the atlas.