Inspiration
The inspiration for HealthPadi came from a simple but powerful truth: access to trusted, timely health information is still a luxury for many people especially in regions where healthcare systems are overwhelmed or inaccessible. We imagined what it would be like if every person had a smart, culturally aware "health friend" that could read the latest medical research, understand their health profile, and warn them before things go wrong.
In Nigeria, “Padi” means “friend” and that’s exactly what we wanted to build: a friendly AI-powered health companion for everyday people.
What it does
HealthPadi is an AI-powered health intelligence app that delivers real-time, personalized medical insights. It continuously monitors global health data, emerging research, local outbreaks, drug recalls, and condition-specific developments—and alerts users if anything relevant to them arises.
Powered by Sonar’s advanced reasoning APIs, HealthPadi doesn’t just answer questions—it explains why the information matters, backs it with citations, and helps users act on it.
How we built it
We used:
- Frontend: React (TypeScript) with Tailwind for clean, mobile-first design
- Backend: Node.js (TypeScript) to handle user profiles, API calls, and Sonar integration
- AI Integration: Sonar API (search, citations, reasoning, deep research)
- Data Sources: Public clinical APIs, CDC/WHO alerts, and user-input health conditions
- Database: MongoDB for structured user data and agent configurations
- Auth & Security: JWT-based auth, Oauth, encrypted user profiles, and privacy-first defaults
Sonar's powerful APIs were integrated as middleware services on our backend to handle prompt creation, API key protection, and tailored response formatting.
Challenges we ran into
- Balancing medical accuracy with accessible language
- Formatting Sonar’s complex reasoning into culturally resonant content
- Optimizing latency for real-time health alerts
- Ensuring privacy while delivering deeply personalized insights
Accomplishments that we're proud of
- Delivering a real-time personalized health alert system
- Creating a culturally resonant UX and brand identity
- Integrating Sonar into a practical, life-changing healthcare solution
- Designing a modular research agent builder for user customization
What we learned
- Simplifying AI-generated medical data requires both empathy and precision
- Prompt engineering is crucial for multi-step health queries
- Users want health tools that feel both intelligent and human
- Cultural relevance is key to adoption and trust in digital health
Built With
- api
- mongodb
- palantir
- react
- socket.io
- sonar
- tailwind
- typescript
Log in or sign up for Devpost to join the conversation.