README.md

---
title: Study RAG Assistant
emoji: 📚
colorFrom: blue
colorTo: indigo
sdk: gradio
app_file: app.py
pinned: false
---

# 📚 Study RAG Assistant (Functional RAG)

This Hugging Face Space runs a **fully functional Retrieval-Augmented Generation (RAG)** system built with:

- Gradio (UI)
- FAISS (vector search)
- Sentence Transformers (embeddings)
- Transformers (LLM generation)
- PDF + Jupyter Notebook parsing

---

## 🚀 What It Does

You can:

- Upload up to 10 `.ipynb` notebooks
- Upload up to 5 `.pdf` files
- Index them into vector embeddings
- Chat grounded strictly in your documents
- Generate structured Notes
- Generate tricky Quiz questions (10–50)
- See citations from retrieved chunks

---

## ⚙️ How It Works

1. Files are parsed and chunked
2. Chunks are embedded using `all-MiniLM-L6-v2`
3. FAISS stores embeddings for fast retrieval
4. Top-k chunks are retrieved per query
5. The LLM generates grounded responses

---

## 📁 Required Files

This Space must contain: