DevPath
Tracks

AI & Modern Engineering

Vectors, embeddings, similarity search, vector databases, LLMs, and RAG.

0 of 10 complete

  1. 01Vectors and EmbeddingsHow text becomes a list of numbers, why distance in that space means similarity, and how to compare vectors.intermediate · 40 min
  2. 02Similarity Search and ANN IndexesWhy exact nearest-neighbour search costs O(n*d), and how IVF clustering and graph indexes like HNSW trade a little recall for a big speedup.intermediate · 45 min
  3. 03Vector DatabasesWhat a vector database adds beyond an array of embeddings: indexing, metadata filtering, upserts, and persistence. Build a tiny store and compare pgvector, dedicated stores, and FAISS.intermediate · 40 min
  4. 04How LLMs Work: Tokens, Context, and SamplingTokens, the context window, next-token prediction, and how temperature and top-p turn scores into text. Includes why models hallucinate and a sampler built from scratch.intermediate · 45 min
  5. 05Prompting and Structured OutputPrompt anatomy, few-shot examples, and how to get reliable JSON from a model: validate every response and retry with the error fed back. Built against a fake llm so it runs offline.intermediate · 40 min
  6. 06RAG: Chunking and RetrievalHow to split documents into chunks, why overlap and metadata matter, and how BM25, dense search, reciprocal rank fusion, and reranking combine into strong retrieval.intermediate · 50 min
  7. 07The RAG Pipeline End to EndIngest, retrieve, augment the prompt, generate, and verify citations: a complete retrieval-augmented generation pipeline you can run offline with stand-ins for the embedding model and the llm.intermediate · 45 min
  8. 08Evaluating RAG SystemsMeasure retrieval with recall@k, MRR, and nDCG, check answers for faithfulness, build a golden set, and learn the common RAG failure modes. All metrics implemented from scratch.advanced · 50 min
  9. 09Agents and Tool UseHow a model calls functions through a loop of think, act, observe: tool schemas, argument validation, step limits and other guardrails, built offline with a fake llm and two tools.advanced · 50 min
  10. 10LLMs in ProductionCost and latency, caching, streaming, rate limits with retries and backoff, guardrails, prompt injection basics, and observability, shown with a small gateway you can run offline.advanced · 50 min