I build robots. Sometimes they work.

SO-101 arms on my living-room table, learned policies, and ordinary household jobs. The interesting question is not merely whether a policy works, but what it actually learned when it does. The setup and what I've learned so far →

Hackathon · 2 days

Toasting bread

Four people, one weekend: bread into the toaster, press the button, retrieve the toast. Three manipulation problems chained together in a demonstration using ACT and SmolVLA.

ACT SmolVLA Deformable Objects

Library robot · Ongoing

Picking and stacking books

In the experiments so far, an ACT policy picked up a few unseen books and a book that was partly occluded.

ACT Informal Trials Occlusion

Desk cleanup · Ongoing

The pen task

I tried several policies on a pen-pickup task. Their behavior changed when I varied the pen, its orientation, or the drop target.

MolmoAct Diffusion Policy VLA

Coursework and research from grad school at Carnegie Mellon and undergrad at Rose-Hulman: computer vision, SLAM, controls, assistive robotics, and CAD.

Graduate · Carnegie Mellon

CMU projects

Robotic post-processing research, lunar rover perception, legged locomotion, and graduate coursework in computer vision and SLAM.

Computer Vision SLAM Motion Planning

Undergraduate · Rose-Hulman

Rose-Hulman projects

A year-long assistive-robotics capstone that represented the USA in South Korea, plus controls, machine-learning, and CAD coursework.

Assistive Robotics Controls CAD