Danjuma is an AI-powered assistant designed to help users with queries related to Moniepoint's services. It utilizes state-of-the-art AI technologies and Retrieval-Augmented Generation (RAG) techniques to provide accurate and contextually relevant responses, leveraging Moniepoint-specific knowledge.

• Intelligent Query Resolution: Uses RAG to retrieve and generate answers based on Moniepoint’s content.
• Chat Interface: A user-friendly interface powered by Streamlit for seamless interaction.
• High-Performance Models: Integrates with Together API for natural language understanding and generation.
• Customizable Knowledge Base: Automatically fetches relevant data using IlimiKudi.

• AI Models: Together API
• Frameworks:
  • txtai for embedding-based retrieval
  • LangChain for workflow orchestration
• Frontend: Streamlit for the chat interface
• Backend:
  • Python for core logic
  • RAG for combining retrieval with generative capabilities

Ensure the following are installed on your system:

• Python 3.8 or higher
• Pip
• Virtualenv (recommended)

1. Clone the repository:

   git clone https://github.com/thelaycon/danjuma.git
   cd danjuma

2. Create a virtual environment:

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

3. Install dependencies:

   pip install -r requirements.txt

4. Run the application:

   streamlit run app.py

5. Open your browser and navigate to http://localhost:8501 to start interacting with Danjuma.

Danjuma utilizes embeddings stored in the moniepoint_index for efficient retrieval and response generation. No additional directories for blog posts or articles are required.

Set the following environment variable as required:

• LLM_API_KEY: Your API key for Together API

• Open the Streamlit interface.
• Enter your query into the chatbox.
• Danjuma retrieves relevant information and generates a response.

• Moniepoint MFB for their content and support
• txtai for embedding-based search
• LangChain for robust AI workflows
• Together API for high-performance LLMs
• IlimiKudi for dynamic data retrieval