Spring 2026 GRASP SFI - Tom Zhang, Daxo Robotics
“From GRASP to Daxo: Rethinking Dexterity in Robotics” ABSTRACT Dexterity remains one of the central unsolved problems in robotics. Many conventional robotic designs may be effective for narrow tasks but are fundamentally limited in the long run for building robots with truly general physical capability. This talk examines why Daxo Robotics has taken a different path: a high-dimensional tendon-driven architecture with 120 actuators in a single hand. It covers the reasoning behind this design, the engineering lessons from building it, and the broader view of robots as systems designed for learning, adaptation, and skill acquisition rather than only task-specific automation. It also reflects on the transition from PhD research to startup building from the perspective of a Penn GRASP alumnus. Alongside the technical story is a discussion of entrepreneurship in robotics: the gap between prototypes and real systems, the tension between conviction and commercialization, and the challenge of building not just a technology, but an enduring company. Presenter Tom Zhang is the founder and CEO of Daxo Robotics, a company building highly dexterous robotic hands for physical intelligence. Daxo’s hand uses more than a hundred actuators in a single hand, making it the highest-dimensional robotic hands ever built, with demonstrated capabilities including real-time continuous handwriting and pen spinning. Tom studied Mechanical Engineering and Computer Science at Cornell University and worked at iRobot, Uber ATG, and Rapyuta Robotics before pursuing a PhD in Computer and Information Science at the University of Pennsylvania’s GRASP Lab. During his PhD, he began his entrepreneurial journey through agricultural robotics, and after graduation founded Daxo’s current effort focused on dexterous robotic hands.

Spring 2026 GRASP SFI - Levi Cai, University of Colorado, Boulder

How to Get Your First 10–20 GoHighLevel Clients in 90 Days | Priya’s POWER Pivot Framework

Spring 2026 GRASP Seminar - Glen Berseth, Université de Montréal

Spring 2026 GRASP SFI - Wei-Hsi Chen, University of Pennsylvania

Andrew Ng: Building Faster with AI

Spring 2026 GRASP on Robotics - Mingmin Zhao, University of Pennsylvania

Future Algorithms 2026 - Session 2 - Dr Pengpeng Hu

Spring 2026 - Robotics Master's Thesis and Capstone Lightning Talks

Sergey Zakharov (Toyota Research Institute): Closing the Loop: Synthetic World for Scalable Learning

AI Agents Full Course 2026: Master Agentic AI (2 Hours)

Spring 2026 GRASP on Robotics - Francesco Bullo, University of California, Santa Barbara

Fall 2024 GRASP on Robotics: Antonio Loquercio, University of Pennsylvania
![Quan Vuong: π0.7, A Generalist Model with Emergent Capabilities [ETHZ Robot Learning 2026]](https://i.ytimg.com/vi/pzolgvyWEFY/hqdefault.jpg?sqp=-oaymwEjCNACELwBSFryq4qpAxUIARUAAAAAGAElAADIQj0AgKJDeAE=&rs=AOn4CLCgAo47f3xoSO6emfV1oRNRgBN_BA)
Quan Vuong: π0.7, A Generalist Model with Emergent Capabilities [ETHZ Robot Learning 2026]

Inside the Mind of Anthropic CEO Dario Amodei | The Circuit | Extended Interview

Robotics' End Game: Nvidia's Jim Fan

Abstract Black and White wave pattern| Height Map Footage| 3 hours Topographic 4k Background

How Proctor’s texts in Karen Read lawsuit could free dangerous criminals

Nvidia CEO Jensen Huang Interview| Bloomberg Technology Special

