Optimizing LiDAR calibration software for mass production with focus on low-latency, algorithmic improvements, and cache-aware concurrent architecture design.

Built LiDAR perception pipeline for autonomous vehicles with optimized 3D object detection and low-latency bounding box inference on constrained compute.

Built C++ perception stack, 0-to-1, real-time data association algorithms, and ROS2 optimization - doubling stack efficiency and detection accuracy.

Working on open-source implementations and 3D visualizations of baseline robotics algorithms for localization, mapping, and controls using PyBullet real-time physics simulation. This research project is under Professor Nikolay Atanasov.

I worked as a software engineering intern and an undergrad research intern for a Research Project in the Autonomy Lab, Contextual Robotics Institute at UC San Diego. The research was on Vision-Guided Quadrupedal Locomotion under Professor Xialong Wang’s group.