My method, FORGE, beat the prior best on humanoid locomotion and standup at 10× less compute per individual.
Agent payment rails got built, but nobody built the trust layer. We're starting with evals and routing for paid tools.
A four-symmetry analysis tied our results to queueing theory and found which symmetry-breaks would change that.
I studied chemical engineering, and in 2023 took third in Zhejiang's provincial physics innovation competition.
Researcher at University of Pennsylvania
I research automated reward design for reinforcement learning and robotics, and I run Cadmus, a marketplace for agents.
My method, FORGE, beat the prior best on humanoid locomotion and standup at 10× less compute per individual.
Agent payment rails got built, but nobody built the trust layer. We're starting with evals and routing for paid tools.
A four-symmetry analysis tied our results to queueing theory and found which symmetry-breaks would change that.
I studied chemical engineering, and in 2023 took third in Zhejiang's provincial physics innovation competition.
Researcher at University of Pennsylvania
I research automated reward design for reinforcement learning and robotics, and I run Cadmus, a marketplace for agents.
My method, FORGE, beat the prior best on humanoid locomotion and standup at 10× less compute per individual.
Agent payment rails got built, but nobody built the trust layer. We're starting with evals and routing for paid tools.
A four-symmetry analysis tied our results to queueing theory and found which symmetry-breaks would change that.
I studied chemical engineering, and in 2023 took third in Zhejiang's provincial physics innovation competition.
By using this service, you agree to the Terms of Service and Privacy Policy.