Inspiration
In today's world, reliable health information is more important than ever, yet it's increasingly difficult to find trustworthy advice. We were inspired to create CareTalk after realizing how many people struggle to access personalized, evidence-based health guidance. The Perplexity API's ability to provide well-sourced information presented a unique opportunity to build an AI health companion that not only offers advice but also cites its sources, bringing transparency and trust to digital health conversations.
What it does
CareTalk is an AI-powered health companion that provides personalized wellness guidance through natural conversations. Users can chat with specialized AI agents focused on different aspects of health, including mental wellness, nutrition, fitness, and sleep. Each response is backed by reliable sources, with direct citations from the Sonar API. The interface is designed to be intuitive, with a clean chat interface that makes complex health information accessible to everyone.
How we built it
- Frontend: Built with Next.js and TypeScript for a responsive, type-safe application
- UI/UX: Styled with Tailwind CSS for a clean, modern interface
- AI Integration: Leveraged the Perplexity Sonar API for accurate, well-sourced responses
- Deployment: Configured deployment on Vercel
Challenges we ran into
- API Integration: Adapting the Perplexity API responses to fit our chat interface while preserving citations
- Streaming Implementation: Implementing smooth streaming of AI responses for a better user experience
- Type Safety: Managing complex TypeScript types across the application
- Responsive Design: Ensuring the chat interface works seamlessly across all device sizes
Accomplishments that we're proud of
- Successfully integrated the Perplexity API to provide well-sourced health information
- Created a beautiful, intuitive UI that makes health information accessible
- Implemented real-time streaming for a more engaging user experience
- Built a system that properly attributes sources, bringing transparency to AI health advice
- Developed a scalable architecture that could easily incorporate more specialized agents in the future
What we learned
- The importance of proper source attribution in health-related AI applications
- How to design conversational interfaces that feel natural and intuitive
What's next for CareTalk
- More Specialized Agents: Expand our team of AI health experts to cover more specific health topics
- User Accounts: Implement user authentication to save conversation history and preferences
- Multilingual Support: Make health guidance accessible in multiple languages
- Voice Interface: Add voice interaction for hands-free health guidance
- Clinical Validation: Partner with healthcare professionals to validate and enhance our AI's advice
We believe CareTalk represents the future of accessible, reliable health information, and we're excited to continue developing it to help even more people take control of their wellness journey.
Built With
- nextjs
- typescript
Log in or sign up for Devpost to join the conversation.