Inspiration

University students often quietly spiral into burnout—missing classes, sleeping erratically, withdrawing from peers. By the time they reach out for help, they're in crisis mode. Campus counseling centers frequently have multi-week waitlists, and the current system only catches students after they break.

The question arose: what if students had a simple, private way to track their own patterns before things became critical? What if counselors could see early warning trends across the student body without violating anyone's privacy?

EarlyMind addresses this gap.

The Idea Early detection of burnout and depression patterns before crisis hits. Instead of waiting for students to collapse, we give them a mirror to see their own patterns forming in real time.

Who We're Building For University students who want to understand their wellbeing privately, and campus health services that need actionable signals without violating student trust.

What It Does

  • Time series trend detection that spots when your sleep, mood, or stress deviates from your personal baseline
  • Quick daily self check-ins (under 30 seconds)
  • Anonymized counselor dashboard showing cohort-level risk patterns—no individual identifiers, ever

Why It Matters The platform is prevention-focused, opt-in, and privacy-first. This isn't therapy. It's not a diagnosis. It's awareness. Students see their data and decide what to do with it. Counselors receive early warning signals without knowing who needs help until students reach out themselves.

From Concept to Code Rather than stopping at mockups, the solution was built directly on GitHub. The student-facing mobile app and counselor dashboard were developed iteratively, translating the core idea into a working prototype through rapid development.

Business Model B2B licensing for universities. Campus health centers pay per student cohort; students use it for free. Add-ons include weekly digest emails for counselors and integration with campus SSO.

Implementation The application uses React with Vite and Tailwind for the frontend. Supabase Postgres with Row Level Security handles anonymized storage. Python-based anomaly detection processes simulated smartphone data (mood scores, sleep hours, screen time). Local processing ensures student data never leaves their device unless they consent.

Privacy & GDPR Compliance

  • Local processing: risk scores calculated on-device
  • Anonymization: UUID-only storage, no names or emails
  • Explicit consent: students must opt in before any data is stored
  • Right to deletion: one-click data erasure via RLS policies

EarlyMind enables earlier intervention, with consent, and without stigma.

Built With

  • ai
  • react
  • supabase
  • tailwind
  • vercel
  • vite
Share this project:

Updates