Inspiration
We wanted to blend the charm of retro Tamagotchis with modern AI and real-world health data. Wellness tools often feel clinical or boring, so we asked: What if your habits directly powered a virtual pet? TamaCare turns everyday choices into visible emotional and energetic changes, creating a fun, nostalgic way to stay mindful about health.
What it does
TamaCare is an AI-powered virtual pet whose mood, animations, and energy depend entirely on your real habits. Log actions through natural language or load your wearable/Health CSV data — your pet reacts instantly. Healthy habits make it thrive; unhealthy habits drain it. Severity is dynamically assessed using Gemini, and the pet displays expressive pixel animations and speech bubbles based on how you're doing.
How we built it
We built a Flask-based backend that will store user state. Currently, it evaluates habits with Gemini, and dynamically updates mood, energy, and streaks. The frontend uses custom pixel-art UI, CRT effects, and animated GIF sprites for a nostalgic Tamagotchi feel. CSV health data (sleep, steps, HRV, exercise, etc.) is parsed and converted into pet stats. We developed a severity-scoring system so AI can quantify how healthy or unhealthy each habit is, and we integrated dynamic animations, transitions, and speech bubbles driven by state changes.
Challenges we ran into
- Getting Gemini to return parsable structured data reliably
- Making animations transition smoothly between emotional states
- Designing a pixel-art UI that feels authentically retro but still readable
- Mapping real health metrics into meaningful pet behavior
- Avoiding sprite pop-out issues and image mismatches across states
- Ensuring the habit scoring system felt “fair” and intuitive ## Accomplishments that we're proud of
- A fully working AI-driven Tamagotchi that responds to real health data
- Custom sprite animations, including dynamic transitions and streak celebrations
- A robust habit scoring system that scales mood/energy based on severity
- Turning a simple idea into a compelling, playful health experience ## What we learned We learned how to build more reliable AI pipelines, how to design responsive pixel-art interfaces, and how complex real health metrics can be when mapping them into behavior. We also gained experience integrating multiple APIs, handling messy CSV data, and balancing game design with wellness logic. ## What's next for TamaCare
- Full Apple Health / Google Fit OAuth integrations
- More pet evolutions, moods, and animations
- A “friend battle” or “friend care” mode where pets interact
- Long-term trend tracking and weekly health reports
- Achievement badges and habit streak boosts
- A mobile app version optimized for daily quick interactionsd
Log in or sign up for Devpost to join the conversation.