Robotics software engineer
I work across localization, computer vision, robot software, and machine learning. This site highlights selected systems, what they do, and how they were built.
A ROS 2 localization pipeline that combines visual SLAM, GPS, IMU, and wheel odometry for autonomous vehicle operation. The system was tested over a two-mile route and designed for debugging in RViz and Foxglove.
Competition robot software for localization, target alignment, autonomous routines, and high-frequency odometry. The system connected perception, motion planning, and mechanism control into one observable workflow.
A modular Java library for reusable robot and vision infrastructure. It organizes state estimation, subsystem interfaces, commands, and telemetry so new robot code can be built faster and maintained more reliably.
An interactive environment for testing robot movement and game-piece behavior before deploying code to hardware. It provides a safer and faster way to explore control logic and physical assumptions.
A multi-task autoencoder and classifier trained on EEG recordings to recognize 30 command categories and map predictions to predefined robot actions.
A real-time perception pipeline that projects road segmentation into bird's-eye view and estimates smooth centerlines for autonomous navigation on embedded hardware.
I am an incoming Computer Science student in the Texas Robotics Honors Program at UT Austin. I enjoy building systems where software must understand the physical world, make decisions, and control real hardware.
My experience includes autonomous vehicles at NASA Johnson Space Center, four years of FRC robotics, and applied machine learning.