Experience

Research Scientist Intern

Meta Reality Labs | Seattle, WA | May-Nov 2026

• Designed a multi-camera 3D eye-localization and eye-tracking prototype combining synchronized imaging, learned segmentation, and geometric reconstruction on embedded hardware.
• Adapted perception models for edge deployment with ONNX and TensorRT, reducing processing latency through GPU processing and profiling.
• Built calibration and synchronized-recording workflows, including a web interface, diagnostics, and recovery from camera failures.
• Designed illumination and control PCBs and microcontroller firmware for synchronization with external motion-capture equipment. Developed automated validation tools and robotic experiments to measure accuracy and repeatability.

Tools: Python, PyTorch, OpenCV, ONNX, TensorRT, NVIDIA Jetson, ESP32/ESP-IDF, GMSL.

Graduate Research Assistant

• Built an open-source, synchronized multimodal data collection system for human-subject studies, combining egocentric video and audio, smartwatch motion data, custom time-of-flight sensing, and medical-device ground truth.
• Led the collection and release of EgoEMS: over 20 hours of data across 233 simulated emergency scenarios with 62 participants, including 46 EMS professionals.
• Co-developed MiDAS, a platform-agnostic multimodal acquisition system for robot-assisted surgery, validated on Raven-II and da Vinci Xi.
• Developed multimodal activity recognition methods for EMS cognitive assistance and real-time surgical activity prediction and safety monitoring.

Tools: Python, PyTorch, C++, CUDA, Azure Kinect SDK, Android, Java, Vuzix AR glasses.