Researcher at World Labs, teaching robots to learn from world models.
I work on robot learning at World Labs. I received my Bachelor's degree in Applied Math and CS from Columbia University, where I was fortunate to be mentored by Yunzhu Li, Antonio Loquercio, and Brian Plancher.
My research sits at the intersection of robotics, computer vision, and machine learning, with a focus on understanding the role of world models in data-driven robotics pipelines to equip robots with more robust, capable, and efficient physical and perceptual capabilities. I aim to use world models to scale robotics progress at the speed of compute rather than real-world clock time.
If you would like to collaborate, feel free to reach out.
TL;DR — We present a shared autonomy framework for reliable teleoperation by learning a residual copilot that provides low-level assistance.
TL;DR — We propose a framework for robot policy evaluation in simulation, using Gaussian Splatting for rendering and a soft-body digital twin for dynamics.
TL;DR — We introduce TAG-K, a lightweight Kaczmarz variant combining greedy row selection and tail averaging for fast online inertial parameter estimation.
TL;DR — We present a well-posed variational formulation and finite element approximation for time-harmonic 2D Maxwell's equations.
