CareCircle

Dashboard
Medications
Appointments
Summaries

Inspiration

Caring for a loved one involves more than just being there for them. Caregivers often have to manage medication schedules, remember appointments, track daily activities, and keep other family members updated. When all this information is scattered across different tools, important details can easily be missed.

We wanted to build CareCircle, an AI-powered caregiving assistant that makes everyday care more organized, connected, and less stressful for both caregivers and the people they care for.

What it does

CareCircle brings essential caregiving tasks together in one place. Its key features include:

Medication reminders: Helps caregivers manage medication schedules and track dose confirmations.
Appointment management: Keeps appointments organized so important visits are easier to remember.
AI-powered daily summaries: Uses Amazon Bedrock to generate concise summaries of caregiving activities, helping caregivers stay informed.
Voice-assisted workflows: Explores Alexa+ integration through a simulator and an MCP-based tool workflow.
Caregiver dashboard: Provides a central interface for viewing medications, appointments, summaries, and alerts.

We also plan to explore optional Ring event integration for relevant delivery or safety notifications.

How we built it

CareCircle is designed around a modular architecture that connects a caregiver-facing dashboard with backend services and AI capabilities.

Frontend: React, Vite, and TypeScript for the caregiver dashboard.
Backend: Node.js, Express, and TypeScript for APIs and application logic.
Database: Amazon DynamoDB is planned for storing caregiving data.
AI: Amazon Bedrock for generating daily summaries and supporting AI-assisted workflows.
Tool integration: Model Context Protocol (MCP) for structured interactions with caregiving tools, alongside an Alexa+ simulation.
AWS deployment: We plan to use services such as AWS Lambda, API Gateway, and Amplify as appropriate for deployment.

We are building the application in stages, starting with the core application foundation and then connecting the caregiving workflows and AI integrations.

Challenges we ran into

One of our biggest challenges was designing a reliable flow between the frontend, backend, database, and AI services. A caregiving application needs more than a good interface: actions such as confirming a medication dose must be handled consistently.

We also had to think carefully about the role of AI in a sensitive domain. AI-generated summaries should communicate recorded information clearly without inventing events or replacing professional medical advice.

Another challenge is integrating voice workflows and backend tools while keeping the system modular, testable, and easy to extend.

Accomplishments that we're proud of

We established a clear architecture for CareCircle and divided the work across frontend development, backend services, AI capabilities, MCP integration, testing, and documentation.

We have also built the initial application foundation, including frontend navigation and backend health, authentication, and user-related modules. These give us a starting point for connecting the core caregiving features.

Most importantly, we have a clear path toward bringing medication management, appointment tracking, and AI-generated caregiver summaries into one connected experience.

What we learned

We learned that building a useful AI application requires more than integrating a language model. Reliable data, clear API contracts, structured tool interactions, and thorough end-to-end testing are just as important.

We also learned to approach AI carefully in healthcare-adjacent applications: AI should help caregivers understand information and coordinate tasks, not make unsupported medical decisions.

Working across frontend, backend, and AI components also reinforced the importance of modular design, clear responsibilities, and testing integrations early.

What's next for CareCircle

Our next steps are to:

Complete the medication, appointment, and dose-tracking APIs.
Connect the frontend dashboard to live backend data instead of mock data.
Integrate Amazon Bedrock for caregiver summaries and validate the results.
Build and test MCP-based workflows for caregiving actions.
Develop and test voice-assisted medication and appointment workflows through the Alexa+ simulator.
Strengthen authentication, data protection, error handling, and end-to-end testing.
Deploy the application on AWS and explore optional Ring event integration.

Our long-term goal is to make CareCircle a dependable caregiving companion that helps families stay organized, informed, and connected while keeping care decisions in human hands.