Projects(hover over images)

Foilpal

Wearable, ML-powered wingfoiling trainer that analyzes motion and gives real-time light cues to improve balance and reduce falls. Integrated an IMU, GPS, LEDs, and a Raspberry Pi to collect movement data and adapt feedback based on prior riding patterns. Visit foilpal.com

Python LSTM Raspberry Pi Sensor fusion
Foilpal electronics enclosure
Wingfoiling with Foilpal
Testing the Foilpal balance board indoors
Foilpal wing and foil board in the water
Foilpal electronics enclosure
Wingfoiling with Foilpal
Testing the Foilpal balance board indoors
Foilpal wing and foil board in the water
Wingfoiling with Foilpal, silhouette

GaitGuard

Co-created a wearable rehabilitation system that compares a user's movement against an ML-generated personalized "healthy twin" and delivers real-time corrective feedback. Recognized at YHack and featured on the official Cornell Bowers CIS website. View on Devpost

Machine learning Wearables Rehab tech
GaitGuard leg sensor
GaitGuard hardware components
GaitGuard team
YHack closing ceremony, overall winners
GaitGuard leg sensor
GaitGuard hardware components
GaitGuard team
YHack closing ceremony, overall winners
GaitGuard live dashboard on laptop and phone
Team celebrating after YHack

GateAgent

A continuation of GaitGuard with more of a research focus, still very much in progress. The idea is to connect GaitGuard's approach with AI agents toward a single, universal program that can independently diagnose and analyze movement, efficiently, hasn't been done before. (Sorry, no pictures yet :( )