ClauseAI is a document processing and querying application that leverages AI-powered tools for extracting metadata, vectorizing content, and providing intelligent query responses. It combines the power of OpenAI models and Qdrant to create a seamless document management system.

Follow these steps to set up and run the ClauseAI project locally:

• Python 3.10 or higher
• Virtual environment manager (optional but recommended)

1. Clone the Repository:

   git clone <repository_url>
   cd ClauseAI

2. Set up a Virtual Environment:

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

3. Install Dependencies:

   pip install -r requirements.txt

4. Set Environment Variables: Create a .env file in the root directory with the following content:

   QDRANT_URL=<your_qdrant_url>
   QDRANT_API_KEY=<your_qdrant_api_key>
   OPENAI_API_KEY=<your_openai_api_key>
   

5. Run the Application:

   streamlit run workflow.py

ClauseAI consists of two main functionalities:

• Upload a PDF document.
• Convert the document to Markdown format and extract metadata.
• Generate vector embeddings for the document content and store them in Qdrant.
• Extract entities using GPT-4 for metadata enrichment.

• Select a processed document by its ID.
• Query the document using two mechanisms:
  • Qdrant: Fetch the most relevant context chunks.
  • LLM: Refine the Qdrant output using OpenAI GPT-4 for a natural-language response.

• PDF to Markdown Conversion: Extracts textual content and metadata from uploaded PDF documents.
• Vectorization: Converts document content into vector embeddings using OpenAI embeddings and stores them in Qdrant for efficient querying.
• Entity Extraction: Uses GPT-4 to identify and extract key entities in the document.
• Intelligent Querying: Combines Qdrant's vector search and GPT-4's natural language understanding to deliver detailed query responses.

We welcome contributions to ClauseAI! Please fork the repository and create a pull request with your changes.