📄 DocChat — RAG Chatbot with Gemini

🚀 Live Demo

👉 Click here to try the app

A production-style Retrieval-Augmented Generation (RAG) pipeline that lets you chat with any PDF document. Built with LangChain, ChromaDB, sentence-transformers, and Gemini 1.5 Flash.

---

🚀 Features

• 📤 Upload any PDF — research papers, reports, notes
• 🔍 Semantic search using all-MiniLM-L6-v2 embeddings
• 🧠 Answers grounded in document context (no hallucination)
• 📍 Page-level source citations on every answer
• 🔎 View retrieved chunks with similarity scores
• 💬 Multi-turn chat history in session

---

🏗️ Architecture

PDF Upload
    ↓
Text Extraction (PyMuPDF)
    ↓
Chunking with overlap (LangChain RecursiveCharacterTextSplitter)
    ↓
Embedding (sentence-transformers: all-MiniLM-L6-v2)
    ↓
Vector Store (NumPy cosine similarity, in-memory)
    ↓
User Question → Embed → Cosine similarity search → Top 5 chunks
    ↓
Prompt = chunks + question → Gemini 1.5 Flash
    ↓
Answer + Page citations

---

📁 Project Structure

rag-chatbot/
├── app.py                  ← Streamlit frontend
├── rag/
│   ├── loader.py           ← PDF text extraction (PyMuPDF)
│   ├── chunker.py          ← Text splitting with overlap
│   ├── embedder.py         ← Embedding + ChromaDB storage
│   ├── retriever.py        ← Semantic similarity search
│   └── generator.py        ← Gemini API call + answer
├── requirements.txt
├── .env.example
├── .gitignore
└── README.md

---

⚙️ Setup & Run

1. Clone the repo

git clone https://github.com/YOUR_USERNAME/rag-chatbot.git
cd rag-chatbot

2. Create virtual environment

python -m venv venv
source venv/bin/activate        # Mac/Linux
venv\Scripts\activate           # Windows

3. Install dependencies

pip install -r requirements.txt

4. Get free Gemini API key

• Go to aistudio.google.com
• Click Get API Key → Create API key
• Copy the key (starts with AIza...)

5. Run the app

streamlit run app.py

Enter your API key in the sidebar, upload a PDF, and start chatting!

---

🌐 Deploy on Streamlit Cloud (Free)

1. Push this repo to GitHub
2. Go to share.streamlit.io
3. Connect your repo → set main file as app.py
4. Under Settings → Secrets, add:

GEMINI_API_KEY = "AIza_your_key_here"

---

🛠️ Tech Stack

Component	Technology
Frontend	Streamlit
PDF Parsing	PyMuPDF (fitz)
Chunking	LangChain RecursiveCharacterTextSplitter
Embeddings	sentence-transformers (all-MiniLM-L6-v2)
Vector DB	NumPy cosine similarity (in-memory)
LLM	Gemini 1.5 Flash (Google AI)

---

👤 Author

Sudeep — GitHub

CSE Graduate · AI/ML Engineer · Building real-world AI systems