iam-veeramalla/RAG-crash-course ? reverse-engineered prompt

Build me a beginner friendly RAG crash course in Jupyter notebooks that teaches how retrieval augmented generation works from the ground up using plain Python.

I want it to walk through the whole process step by step, starting with why normal LLMs fail on private documents, then showing how to call an LLM, create embeddings, split text into chunks, search for the most relevant pieces, and finally combine everything into a working question answering system with sources. Use a small fictional company document set so the answers feel realistic and clearly come from the files, not from the model’s memory.

Please keep it easy to follow for someone with basic Python only, and include a simple setup guide plus an example environment file. If it makes sense, add a final notebook that shows the same pipeline using a framework so I can see how it simplifies the code. Look up current docs online if you need to.