Robotics · Tactile Sensing · Computational Design
I am a PhD student in Electrical and Computer Engineering at Carnegie Mellon University, where I am a member of the Laboratory of Computational Invention (LOCI), advised by Prof. Andrew Spielberg. My research lies at the intersection of robot learning, tactile sensing, computational design, and advanced manufacturing. I aim to develop physically intelligent robots that learn to perceive, reason, and act through contact, with a particular focus on dexterous manipulation, generalizable robot skills, and the co-design of robotic hardware and learning systems.
I am currently a Research Resident at 1X Labs, where I work on next-generation robot hands and full-stack tactile sensing systems for dexterous manipulation. My research explores how sensor-rich robotic bodies can generate physically grounded data and representations that enable robots to learn more effectively from interaction with the world.
Previously, I conducted research at MIT CSAIL with Prof. Wojciech Matusik and Prof. Daniela Rus. My work has been published in Nature and at robotics venues including CoRL, RSS, ICRA, and RoboSoft. I have also worked at the Robotics and AI Institute on computationally designed, multi-stiffness robot skins and spent several years at Inkbit developing advanced additive manufacturing technologies and robotic applications. I earned a bachelor’s degree in mechanical engineering from Wentworth Institute of Technology.
In my free time, I enjoy long-distance cycling, competitive inline skating, designing miniature ecosystems, and experimenting with fermentation.
Selected work
I work across soft robotics, tactile sensing, robotic assembly, and computational design and manufacturing. My research combines simulation, learning, and hands-on fabrication to create robots that can sense, assemble, and adapt.
NeuralActuator: Neural Actuation Modeling for Robot Dynamics and External Force Perception
A differentiable actuator model for low-cost robots that supports dynamics prediction, sensorless force perception, and force-aware real-robot control.
Printed Helicoids with Embedded Air Channels Make Sensorized Segments for Soft Continuum Robots
Embedded pneumatic channels turn 3D-printed helicoid structures into sensorized segments for a meter-scale, 14-DoF soft continuum robot.
Fabrica: Dual-Arm Assembly of General Multi-Part Objects via Integrated Planning and Learning
A dual-arm system that plans and executes long-horizon, contact-rich assembly of complex objects without human demonstrations or task-specific domain knowledge.
Vision-Controlled Jetting for Composite Systems and Robots
Closed-loop vision enables contactless multi-material jetting of high-resolution soft robots, including a tendon-driven hand, walking robot, and heart-like pump. *Main point of contact for more than two years.
Additional publications
Towards Generalizable and Adaptive Multi-Part Robotic Assembly
A planning-and-learning system for generalizing multi-part robotic assembly to unseen object geometries, assembly paths, and grasp poses.
ASAP: Automated Sequence Planning for Complex Robotic Assembly with Physical Feasibility
Automated sequence and motion planning for physically feasible, multi-step robotic assembly of complex objects.
Selected roles
Recognition
Community