RAG Chatbot

This repository implements a Retrieval‑Augmented Generation (RAG) chatbot in Python, following the step‑by‑step construction guide from python-rag/BUILD-A-RAG-PYTHON.md. It will attempt to use comunity resources and write ups that are stored in the repo. Note: I do not own them. The context documentation there is for learning purposes. Originals where made public by the authors and are available in their web sites and/or git repos. Uncited ones were made by me inspired by the testing context/knowledge.

Project structure

• src/ – application modules
• tests/ – unit tests
• data/raw/ – raw scraped HTML files
• data/processed/ – JSON documents (text, html, etc.)
• eval/golden.py – golden questions used for testing
• scripts/ – helper scripts (scrape, process, run)

Getting started

python -m venv .venv
source .venv/bin/activate  # On Windows use `.venv\Scripts\activate`
pip install -r requirements.txt

Start JupyterLab for interactive exploration:

jupyter lab

Run the Streamlit UI (after scraping and building the index):

streamlit run app.py

Follow the phases described in the guide to scrape, chunk, embed, retrieve and generate.