I built VisionSense as a facial recognition web application designed for secure, real-time detection and authentication. The project’s goal was to explore how computer vision could enhance login systems and accessibility without compromising performance.
The app includes features like secure authentication flows, real-time facial recognition with 90% accuracy, and responsive UI across devices. I integrated Redux for state management and Bootstrap for a clean design system. Continuous deployment on Vercel ensured 99% uptime while Jest and React Testing Library helped achieve 95% test coverage, boosting reliability in production.