MS Computer Science @ UC San Diego. I build machine learning systems that connect perception, simulation, and real-world applications, with a focus on 3D motion reconstruction and intelligent data pipelines.
View ProjectsBuilt an end-to-end pipeline to reconstruct 3D quadruped motion from video using segmentation, skeletonization, and SMAL model fitting. The system transforms raw video into structured joint trajectories for simulation and analysis. Integrated the motion into a Unity-based animation system to generate realistic synthetic data for radar training.
Developed a real-time machine learning system for classifying gamma-ray waveforms using a CNN with uncertainty calibration. Engineered temporal features and optimized inference to achieve low-latency streaming performance. Integrated the model with Unity and HoloLens to visualize radiation signals in an interactive 3D environment.