Inspiration
PDXCareRelay was inspired by the difficulty frontline workers face when trying to help someone in crisis using outdated, scattered, or hard-to-use resource information. We wanted to build a tool that helps workers make faster, clearer decisions in the moment.
What it does
PDXCareRelay helps frontline workers in Portland find services, log referrals, and track the next step. Users can search for providers, filter by need, sort by ZIP code distance, see who is accepting now, get AI-powered suggested options, and save referrals with notes, urgency, transportation, and follow-up details.
How we built it
We built PDXCareRelay using React, Tailwind CSS, Node.js, Express, MongoDB Atlas, Socket.IO, and the Anthropic API. The frontend supports resource search, filters, referral logging, and referral tracking. The backend stores resources and referrals and powers the AI suggestion flow.
Challenges we ran into
Our biggest challenge was making the app feel like one connected workflow instead of separate features. We also had to make sure the AI suggestions were grounded in live local resource data instead of generic output.
Accomplishments that we're proud of
We are proud that PDXCareRelay goes beyond a static directory. It supports the full workflow of finding a service, taking action, and tracking what happens next. We are especially proud of the Suggested Next Step feature and the referral management view.
What we learned
We learned a lot about full-stack integration, state management, MongoDB data modeling, Socket.IO, and designing practical AI-assisted workflows. We also learned how important it is to build around real user needs.
What's next for PDXCareRelay
Next, we want to add provider-side updates, stronger recommendation logic, map support, role-based access, and better follow-up tools so the app can support even more complete care coordination.
Built With
- and
- anthropic
- built-with-javascript
- express.js
- mongodb-atlas
- node.js
- react
- socket.io
- tailwind-css
- the
Log in or sign up for Devpost to join the conversation.