Table of Contents

• Table of Contents
• Installation
• Environment Variables
• Problem Description
• Proposed solution
• System diagram

Installation

• Clone the repository using Git

  git clone https://github.com/SenZmaKi/Kenyare.git
  cd Kenyare

• Install Poppler 24.08.0 or higher and add it to PATH

  • Linux/Mac

    • Install Poppler using your package manager

  • Windows

    • Download Poppler from this link
    • Extract the downloaded file
    • Add the bin folder Release-24.08.0-0\poppler-24.08.0\Library\bin\ to PATH

• Install Python 3.12.4 or higher then run

  • Create virtual environment

    python -m venv .venv

  • Activate virtual environment

    • Linux/Mac

      source .venv/bin/activate

    • Windows

      .venv\Scripts\activate

  • Install dependencies

    pip install -r kenyare/requirements.txt

  • Run the backend api server

    python -m kenyare.server

    By default the server will run on port 8000. You can change the port by setting the FLASK_PORT environment variable. By default the server will run on host 127.0.0.1. You can change the host by setting the FLASK_HOST environment variable.

• Install Node.js 22.9.0 or higher then run

  • Install dependencies

    npm install

  • Run the frontend server

    • Development

      npm run dev

      Navigate to localhost:5173 on your browser. You can change the port and host by setting the VITE_DEV_PORT and VITE_DEV_HOST environment variables.

    • Production

      npm run build
      npm run preview

      Navigate to localhost:4173 on your browser. You can change the port and host by setting the VITE_PROD_PORT and VITE_PROD_PORT environment variables.

    By default the frontend server will make backend api requests to localhost:8000 set the FLASK_PORT and FLASK_HOST environment variables to change the port and host.

Environment Variables

• Create a .env in the root project directory. The environment variables will be automatically loaded.

OPENAI_API_KEY=sk-proj-xxxxxxxxxxxxxxxxx # Required
FLASK_PORT=8000
FLASK_HOST=127.0.0.1
VITE_DEV_HOST=127.0.0.1
VITE_DEV_PORT=5173
VITE_PROD_HOST=127.0.0.1
VITE_PROD_PORT=4173
CLEAR_UPLOADS_DIR=0 # Set to 1 to clear upload directory on first quotation input run
CLEAR_QUOTATATIONS_DIR=0 # Set to 1 to clear quotations directory on backend server start
DELETE_UPLOADS=0 # Set to 1 to delete uploaded files after extracting quotation input

Problem Description

The current process of handling Professional Indemnity Insurance (PII) quotation requests is largely manual, involving the analysis of proposal forms and associated documents. This approach is time-consuming and labor-intensive, resulting in significant delays in the submission of quotes to our business partners. Presently, we rely on an Excel template for data entry and quote generation, which not only increases the likelihood of errors but also hampers efficiency and responsiveness.

As the insurance industry evolves and the demand for timely and accurate quotes grows, our manual processes are becoming a bottleneck. The inability to quickly analyze and process requests undermines our operational effectiveness and customer satisfaction. To maintain our competitive edge and ensure optimal service delivery, there is an urgent need to automate the reinsurance underwriting quotation process. This transition aims to enhance efficiency, accuracy, and overall value for our business partners, ultimately driving better outcomes for all stakeholders involved.

Proposed solution

The proposed solution is an AI-powered bot designed to automate the processing of Professional Indemnity Insurance (PII) quotation requests. This bot will autonomously handle the majority of the process, minimizing the need for manual input and significantly reducing time delays.

System diagram