AI Knowledge Worker

An Expert Question-Answering Agent for Insurance Tech

This project is an AI-powered question-answering agent designed to assist employees of Insurellm, an Insurance Tech company. Built using Retrieval-Augmented Generation (RAG), the system ensures accurate and cost-effective responses by leveraging vector-based document retrieval and OpenAI's LLMs.

---

Table of Contents

1. Features
2. Technologies Used
3. Folder Structure
4. Installation
5. Usage
6. Deployment
7. Contributing
8. License

---

Features

• Retrieval-Augmented Generation using LangChain
• Vector embedding storage using ChromaDB
• Interactive chatbot interface powered by Gradio
• Secure integration with OpenAI API
• Supports text documents from the knowledge base
• Visualizes vector embeddings using t-SNE

---

Technologies Used

• Python 3.9
• LangChain (Document loading, vector retrieval, and LLM integration)
• OpenAI GPT-4o-mini (LLM for question answering)
• ChromaDB (Vector database for document embeddings)
• Gradio (Chatbot UI)
• Matplotlib and Plotly (Vector visualization)

---

Folder Structure

AI_worker  
├── app.py               # Main application script  
├── knowledge-base       # Folder containing knowledge documents  
├── vector_db            # Folder for vector database storage  
├── requirements.txt     # Required dependencies  
└── README.md            # Project documentation  
Installation

To run this project locally, follow these steps:

    Clone the repository:

git clone https://github.com/yourusername/AI_knowledge_Worker.git  
cd AI_knowledge_Worker  

    Set up a virtual environment (recommended):

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

    Install dependencies:

pip install -r requirements.txt  

    Set up environment variables:
    Create a .env file in the root directory and add your OpenAI API key:

OPENAI_API_KEY=your_openai_api_key_here  

Usage

    Build the vector database (if not already created):

python app.py  

    Launch the chatbot interface using Gradio:

python app.py  

The interface will be available at:
http://localhost:7860
Deployment
Deploy on Hugging Face Spaces (Recommended)

    Upload the entire AI_worker folder to Hugging Face Spaces.
    Set the OPENAI_API_KEY under Settings > Secrets in Hugging Face.
    The app will automatically build and deploy, providing a public URL for access.

Deploy Using Docker

    Build the Docker image:

docker build -t ai-knowledge-worker .  

    Run the container:

docker run -p 7860:7860 -e OPENAI_API_KEY=your_openai_api_key_here ai-knowledge-worker  

Contributing

Contributions are welcome. Feel free to submit issues and pull requests to improve the project.

    Fork the repository
    Create a new branch:

git checkout -b feature-branch  

    Commit your changes:

git commit -m 'Add new feature'  

    Push to the branch:

git push origin feature-branch  

    Submit a pull request

License

This project is licensed under the MIT License.
Contact

For inquiries or collaboration opportunities, please reach out via LinkedIn or email.