RAG & Vector Databases
Make models answer from your data, not their guesses.
Build a real Retrieval-Augmented Generation pipeline. Chunking, embeddings, vector stores, hybrid search, evaluation — and the failure modes nobody warns you about.
8h
Duration
10
Lessons
5.0k
Learners
Path map
Lessons unlock as you complete the previous one. Your progress is saved on this device.
Lesson 1
What RAG actually is — and when not to use it
9m35 XPLesson 2
Cosine similarity in 5 lines of code
9m35 XPLesson 3
Chunking — the most important boring decision
11m45 XPLesson 4
Vector databases — what they actually do
10m40 XPLesson 5
Build a tiny end-to-end RAG
13m55 XPLesson 6
Hybrid search & rerankers
10m40 XPLesson 7
The five most common RAG failure modes
10m40 XPLesson 8
Evaluating a RAG pipeline
12m50 XPLesson 9
RAG in production: cost, latency, freshness
10m40 XPLesson 10
Capstone — design RAG for support tickets
15m80 XP