How iam-veeramalla/rag-crash-course works
RAG Crash Course is a notebook-based teaching product: learners build a document question-answering system step by step, then compare it with a LangChain version. The documented system uses sample company documents, chunking, embeddings, vector search, and an LLM to produce sourced answers. The README establishes the lesson sequence and broad pipeline, but notebook source was not sampled, so internal calls and detailed data transfers cannot be verified.
Main entry points: 06-build-a-rag-pipeline.ipynb RAG pipeline · 01-what-is-rag.ipynb RAG basics · 02-calling-an-llm.ipynb LLM calls · 03-embeddings.ipynb Embeddings
Components 11
Course notebooks
- RAG basics
notebooks/01-what-is-rag.ipynb - LLM calls
notebooks/02-calling-an-llm.ipynb - Embeddings
notebooks/03-embeddings.ipynb - Document chunking
notebooks/04-chunking.ipynb - Vector search
notebooks/05-vector-search.ipynb - RAG pipeline
notebooks/06-build-a-rag-pipeline.ipynb - LangChain version
notebooks/07-rag-with-langchain.ipynb
Knowledge source
- Company documents
data
External services
- OpenAI models
- ChromaDB
Other components
- Learner
Connections 10
- Learner → RAG basics: starts course
README.md - RAG basics → LLM calls: continues to
README.md - LLM calls → Embeddings: continues to
README.md - Embeddings → Document chunking: continues to
README.md - Document chunking → Vector search: continues to
README.md - Vector search → RAG pipeline: continues to
README.md - RAG pipeline → LangChain version: compares framework
README.md - RAG pipeline → Company documents: answers questions over
README.md - RAG pipeline → OpenAI models: uses models
README.md - RAG pipeline → ChromaDB: uses vector search
README.md