💡 Inspiration As student-athletes, we realized that managing wellness is a daily balancing act — juggling early practices, late-night studying, inconsistent sleep, and rushed meals. We noticed that while there are plenty of fitness or sleep apps, none provide a unified, personalized system built for athletes who need structured guidance across sleep, diet, and exercise. That gap inspired us to create Snore.AI — a centralized wellness companion designed to help athletes rest smarter, eat better, and perform stronger.

What It Does Snore.ai is an AI powered web app that helps student-atheletes understasnd and improve their wellness through three key areas: 💤 Sleep: Predicts and recommends optimal bedtime, wake time, and duration based on lifestyle factors like caffeine, stress, and screen time. 🥗 Diet: Suggests meals and nutrients tailored to each athlete’s sport, body needs, and recovery cycle. 🏋️ Exercise: Generates workout guidance aligned with energy levels, recovery, and daily goals. Everything is connected in one dashboard, creating a clear, data-driven picture of overall wellness.

🛠️ How We Built It We built Snore.AI using a modern full-stack architecture: Frontend: Next.js (React + TypeScript), styled with Tailwind CSS for responsive, clean design. Backend: Node.js with Express and Prisma ORM connected to a PostgreSQL database. AI Layer: TensorFlow.js powers our Sleep Optimization Model, which predicts ideal sleep schedules using personalized data. Integrations: Google Calendar for scheduling, with future plans for Fitbit and Apple Health APIs.

🚧 Challenges We Ran Into Designing a realistic sleep model that combined behavioral data (like caffeine and stress) with biological timing. Balancing AI explainability — ensuring that predictions make sense and feel actionable to users. Getting data normalization right for time-based features (like bedtime and wake time). Coordinating frontend–backend–AI communication seamlessly while staying hackathon-fast.

🏆 Accomplishments That We’re Proud Of Built a fully functional AI-driven sleep optimization model using TensorFlow.js. Created a responsive, athlete-focused web interface that feels modern and intuitive. Designed an architecture that allows for cross-domain integration (sleep, diet, exercise) in one system. Delivered an MVP with real predictions and clear, personalized recommendations — all in under a hackathon weekend.

Built With

Share this project:

Updates