Inspiration
Movement quality isn’t just important for athletes—it affects everyone. From people working out at home to those recovering from injuries, most lack access to real-time feedback. Coaching and physical therapy can be expensive or inaccessible.
We were inspired to build a tool that #democratizes movement coaching, making high-quality feedback available to anyone, anytime.
What it does
Kinetix is an intelligent movement coach that analyzes exercises and provides #real-time, actionable feedback.
It detects unsafe or inefficient form, flags potential injury risks, and gives clear coaching cues so users can improve while training independently.
Kinetix focuses on movement quality, not just motion tracking.
How we built it
Pose Estimation & Analysis: We use MediaPipe Pose to detect 33 body landmarks in real time, which we then process to compute joint angles and analyze movement patterns.
Gemini API: We used Gemini to help us analyze the pose estimation.
Google AI Studio: Used to prototype and refine feedback systems.
Frontend (Next.js): Created an interface for live tracking and video uploads.
Figma: Designed a clean, modern UI with a dark, high-contrast style.
Challenges we ran into
Our biggest challenge was turning ambitious ideas into a working product under time constraints.
We had to balance creativity with feasibility while handling noisy pose data, reducing false positives, and integrating real-time feedback into the UI.
Accomplishments that we're proud of
We’re proud of building a system that delivers #meaningful movement feedback, not just pose tracking.
We successfully:
Built real-time analysis Created a scoring + feedback system Designed an intuitive UI What we learned
We learned how to:
Work with real-time, imperfect data Break down complex systems Collaborate effectively under pressure
Most importantly, we learned how to turn ideas into #real, usable products.
What's next for Kinetix
We plan to expand Kinetix into a #digital movement coaching platform with:
Reference-based comparisons Personalized analysis Injury risk prediction Broader use cases (rehab, fitness, yoga)
Our goal is to help people #move better and stay injury-free.
Log in or sign up for Devpost to join the conversation.