CareBridge

Logo
Dashboard

Inspiration

Our team member Nazeer, realized during his time working as a caretaker that there are no systems in place to review and assign caretakers to the clients that require assistance with their activities of daily living such as ambulation, cooking, etc. This is a rising problem in the United States, for example the number of family caretakers helping older adults rose from 18.2 million in 2011 to 24.1 million in 2022 (a ~32% increase).

What it does

Our project is a platform where both caretakers and clients looking for assistance (elderly, chronic illness, veterans, etc) can find the best fit for each other. After each party creates an account and fills out their respective form, clients can use our AI matching feature to find the best caretaker for them. The AI evaluates the different factors (taken from the forms) and ranks each caretaker based on how well they would fit the client, allowing for a robust and comprehensive matching-making system.

How we built it

The app utilizes Python and Flask for the backend server and Psycopg2 to communicate with our PostgreSQL database. We use our database to store user information for signin, client, and caregiver data. This data is utilized by the Google Gemini ADK, which uses the Loop Agent Protocol to determine the compatibility between caretakers and clients. The cherry on top is ElevenLabs, which we used to improve accessibility for our elderly patients through narration and dictation integration.

Challenges we ran into

The most significant challenge we ran into was finding a relevant use case for adding google gemini into our project and then implementing it to match qualitative and quantitative data to create a matching system. Also working in a team of developers with all different OSs presented a challenge throughout the development cycle.

Accomplishments that we're proud of

We are extremely proud of creating an end to end application with a stunningUI design and a novel implementation of AI features. Adding accessibility for users with impairments also provided us a sense of accomplishment.

What we learned

During our time working on the project, we learned about database design, full stack development, working in a group with other developers, and implementation of AI apis.

What's next for CareBridge

Some next steps include deploying the project on render, adding more features such as google maps apis to determine geolocations and distances for caretakers to limit max appointment distances, booking appointments and payment processing using apis such as stripe. Other great to have features would be adding an urgent help feature for same day appointments, in app messaging, and a search bar to filter based on specific criteria such as languages and services provided.