CareRelay

CareRelay
CareRelay | AI-Powered Family Coordination with Amazon Bedrock
Main CareRelay dashboard
Shared check-in 1
Shared check-in 2
Private check-in 1
Private check-in 2

Inspiration

Families often need to coordinate care, share updates, and remember follow-up tasks, but important information can become scattered across conversations. At the same time, not every personal update should automatically be shared with everyone.

This inspired us to build CareRelay, a consent-first family coordination app that helps families stay informed while respecting individual privacy.

What it does

CareRelay allows a user to record a check-in and choose whether it can be shared with their family. When sharing is permitted, the app can generate a concise family brief and organize follow-up information into a clearer plan. When a check-in is marked private, CareRelay does not create a family brief from that check-in.

How we built it

We built CareRelay as a web application with a responsive dashboard and a backend API. We used Amazon Bedrock with Amazon Nova 2 Lite to generate AI-powered family briefs. We deployed the application on Render and used GitHub for source control and project documentation.

A key part of the implementation is the consent check: the private-check-in flow prevents family-brief generation, rather than relying only on a message asking users to respect privacy.

Challenges we faced

One of our main challenges was connecting the AI service to the deployed application and configuring AWS Bedrock access correctly. We also worked on making the interface easier to understand and ensuring that the private and shared workflows behave differently.

What we learned

Building CareRelay taught us how to connect a web application to a hosted AI service, configure cloud credentials, deploy a backend, and design a user experience around consent. We also learned that AI features should be designed with clear boundaries, not added without considering privacy.

What's next

We want to improve task assignment, add clearer consent controls, strengthen testing around privacy boundaries, and make family coordination more useful while keeping users in control of what they share.