The Challenge
Medicare plan shopping is overwhelming for seniors. They face dozens of plans with complex pricing structures, confusing jargon, and critical decisions about their healthcare. I asked: could I learn from the best existing designs and rapidly build something comparable using AI?
The Approach
Rather than starting from scratch, I conducted a design reconnaissance exercise on Blue Cross Blue Shield of Massachusetts's Medicare shopping experience. The goal wasn't to copy; it was to extract the patterns that work: large readable fonts, clear price displays, and limited comparison options to reduce cognitive overload.
Working with Claude, I documented everything systematically: visual design observations, database schemas for complex plan data, senior-friendly UX patterns meeting accessibility standards, and security requirements for healthcare applications.
The Outcome
The result was a comprehensive prompt that enabled Claude Code to generate a complete, production-ready Medicare plan shopping application. The generated code includes React components, TypeScript interfaces, mock data structures, and a service layer ready for real API integration.
The Bigger Picture
This project demonstrates two viable paths forward: approaching insurance carriers with a turnkey modernization capability, or building an independent platform using public CMS.gov data to serve Medicare beneficiaries nationwide.
What took traditional teams months of discovery and development now happens in hours, with AI handling the heavy lifting while humans provide the strategic direction.