Module 18

RAG and agent

embedding, chunking, vector search, RAG, agent loop

When you finish this module you can say:

I can turn text into embedding vectors, find the pieces closest in meaning to a question, hand those to a language model as context to build RAG, and write a simple agent loop.

By the end of this module you'll be able to write a RAG assistant that answers questions about your own notes by finding the relevant pieces first, and a simple agent loop that can call a tool over and over.

Lessons

6 lessons · ≈ 60 min

1. Embedding and cosine similarity9 min
2. Chunking9 min
3. Vector search10 min
4. The RAG pipeline10 min
5. Agent loop and evaluation10 min
6. Project: An assistant for your own notes12 min · project

Practice exercises

Short writing exercises, independent of the lessons: Go to the exercises

1. An embedding that counts words
2. Build a vocabulary
3. Find the most similar sentence