Skip to content

MyTwin

Team: 404 Brain Not Found (Miloh Zwahlen, Keith Mader)

Case: My Future Health - Meet Your Digital Twin

Platform: Android

Project Type: Personal health twin

Event: START Hack Tour 2026 Interlaken

What is MyTwin?

Most health apps show isolated numbers. They tell you how many steps you took, how long you slept, or what your heart rate is — but they don't turn those signals into something personal, explainable, and emotionally tangible.

MyTwin turns your recent health behavior into a living digital twin:

  • It pulls wearable data from Samsung Health
  • Combines it with your profile and habit data
  • Simulates short-term and longer-term health trajectories
  • Lets you chat with an AI version of yourself

The result is a health experience that feels less like a dashboard and more like a conversation with your future self.

Key Features

  • Samsung Health Integration

    Reads recent and historical wearable data: sleep, steps, heart rate, and stress proxy — directly from Samsung Health.

  • Manual Overrides

    On the Home screen you can override any metric for today. Wearable data delayed? Missing? Wrong? You stay in control.

  • 7-Day Health View

    Browse the last 7 days of health data. Historical snapshots are read-only; today's values support live override.

  • Rule-Based Simulation

    An offline, explainable simulation engine produces scores for Recovery, Energy, Focus, Strain, and Long-Term Risk — plus what-if scenario comparisons.

  • AI Twin Chat

    Real streaming chat powered by Claude. The AI speaks in a digital twin voice built around your anonymized health context and precomputed simulations.

  • Food Scanner (coming soon)

    Placeholder screen for a future Open Food Facts barcode integration.

Philosophy

MyTwin is not a clinical decision engine. It is a research-driven, interactive prototype for preventive health reflection, behavior change motivation, and future-self exploration.

The digital twin concept becomes compelling in a hackathon demo not by showing data, but by making future consequences feel personal.