Insurance Risk Assessment App

Insurance Risk Assessment

Problem

Re-insurance underwriters at Hannover Re needed to compute health risk scores from applicant data to evaluate policies. Their existing tools were server-based legacy calculators -- slow to update, difficult to maintain, and inaccessible on modern devices. The underwriting team needed a faster, web-based solution they could use from anywhere, with support for multiple languages to serve their global operations.

Approach

I worked on a small team -- two developers, a QA engineer, and a product manager -- to modernize these tools. My first task was migrating the legacy server-based calculators into standalone, client-side applications. I rebuilt the calculation engine in JavaScript, pulling reference data from AWS S3 rather than relying on server-side processing.

Since this was a global product, I added multi-language support. I also set up authentication through AWS Amplify and built a responsive frontend with React and Bootstrap so underwriters could access the tool from any device.

Technical Stack

  • React
  • JavaScript
  • AWS Lambdas
  • AWS S3
  • AWS Amplify
  • React-Bootstrap

Solution

A web-based risk assessment calculator that computes health indicators from user inputs for re-insurance underwriting. Underwriters enter applicant health data and receive computed risk scores instantly in the browser, with no server round-trips for the core calculations. The app supports multiple languages, authenticates users through AWS Amplify, and works responsively across desktop and mobile. It replaced the legacy system with something faster, more maintainable, and accessible from anywhere.

My Role

Frontend developer responsible for the calculation engine, UI implementation, and legacy migration. I wrote the JavaScript functions that power the risk scoring, built the responsive React interface, integrated AWS services for authentication and data storage, and implemented the multi-language support. I also handled the migration path from server-based legacy code to standalone client-side applications.