About the Project

Inspiration

Most fitness apps assume that more effort leads to better results. From personal experience and sports science research, we realized that performance is often limited by recovery, not motivation. Wearable devices already collect HRV and sleep data, but these signals are rarely explained in a meaningful way.

We wanted to build a system that treats recovery as the starting point for training decisions — helping users understand when to push and when to rest, and why.


What We Built & Learned

We built a Recovery-Based Workout Recommender that analyzes wearable data to classify daily recovery as HIGH, MODERATE, or LOW based on:

  • HRV relative to a personal baseline
  • Recent HRV trends
  • Sleep duration

Instead of using black-box scores, we focused on simple, transparent logic and clear explanations. This taught us that recovery signals are highly individual, and explainability is just as important as accuracy.


Challenges

The main challenges were handling noisy wearable data, choosing interpretable logic over complex models, and keeping the project focused within hackathon time constraints.


Takeaway

This project reinforced a simple idea: recovery decides performance. By grounding training decisions in personal recovery data, we help users train smarter — not just harder.

Built With

Share this project:

Updates