✧ EyeCraft

Ranked first in entertainment track for HackMIT among 350+ projects created by over 1,000 undergraduate hackers.

Developed a Minecraft mod using Fabric API, enabling handicap-accessible gameplay through alternative input methods.

Integrated a live facial recognition pipeline using MediaPipe and OpenCV, mapping head poses, facial expressions, and voice commands to real-time game controls.

Trained neural networks on facial data with TensorFlow, achieving over 90% validation accuracy for gesture recognition.