📄 Fine Print Decoder

A simple AI-based application that helps users understand PDF documents by answering questions using Retrieval-Augmented Generation (RAG).

The system retrieves relevant content from the document and provides answers along with Section and Clause references.

🚀 Features

• Upload any PDF document
• Ask questions in natural language
• Retrieves relevant content using FAISS
• Generates answers using a local LLM (Ollama - LLaMA3)
• Displays Section, Clause, and snippet for each answer
• Maintains chat history

🧠 Tech Stack

• Python
• Streamlit
• LangChain
• FAISS
• Ollama (LLaMA 3)

📁 Project Structure

fine-print-decoder/ │── app.py # Streamlit UI │── rag.py # RAG logic │── requirements.txt │── README.md

⚙️ Setup

1. Clone the repository:

git clone cd fine-print-decoder

2. Create virtual environment:

python -m venv venv venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Install Ollama and run model:

ollama pull llama3 ollama run llama3

5. Run the app:

streamlit run app.py