🚀 Insurance Assistant

Insurance Assistant is a full-stack AI-powered solution designed to assist users with document analysis and search using vector embeddings and natural language processing.

PPT, Demo and Behind the Scene: https://drive.google.com/drive/folders/1SLc2uGkSp3JZPcQDaWbybovcQuUNrqAJ?usp=sharing

---

🧰 Tech Stack

Backend:

• Python 3.10+
• FastAPI
• LangChain
• FAISS (Vector Store)
• PyPDF2
• Tiktoken
• Uvicorn

Frontend:

• HTML
• CSS
• JavaScript

---

📦 Folder Structure

hackrx/
├── app/               # Backend app logic
├── data/              # Uploaded documents
├── frontend/          # Frontend React application
├── prompts/           # Prompt templates
├── utils/             # Utility functions
├── vector_store/      # FAISS vector storage
├── index_documents.py # Script to index PDFs
├── run.py             # Launch FastAPI server
├── requirements.txt   # Python dependencies

---

🛠️ How to Run the Project

🔧 Backend Setup

1. Create and activate a virtual environment

   python -m venv venv
   source venv/bin/activate  # For Windows: venv\Scripts\activate

2. Install dependencies

   pip install -r requirements.txt

3. Index your documents

   python index_documents.py

4. Start the backend server

   uvicorn run:app --reload

---

🌐 Usage

• Upload PDF files via the frontend interface.
• Ask questions or search content from the documents using natural language.
• The system uses FAISS and LangChain to return semantically relevant answers.

---

👥 Team Members

• Vijay Ramdev - [email protected]
• Yuvraj Singh - [email protected]
• Diya Agarwal - [email protected]
• Dinesh Patel - [email protected]
• Aayush Laddha - [email protected]

---

📄 License

This project is open-source and available under the MIT License.