Sim2World
I lead an initiative to connect policies developed in simulation with evaluation on physical robots.

IndustrialNext CTO. PhD in AI & Robotics. x-Founder (acquired). Building Robots for Everyone.
I’m an entrepreneur, roboticist, and AI researcher working toward one lifelong mission: free humanity from manual labor.
As CTO of Industrial Next, I build Physical AI and Real2Sim2Real systems for general-purpose robots, starting with manufacturing. Previously, I founded Zordi, where we built AI-powered robotic systems for agriculture before the company was acquired by Industrial Next.
My work sits at the intersection of robotics, machine learning, and real-world deployment, with a focus on dexterous manipulation, robot learning, foundation models, and the systems required to make intelligent robots work reliably in the physical world.
As part of that work, I lead Sim2World, an open benchmarking initiative connecting simulation-based robot learning with real-world evaluation. We’re building a shared testing ground where researchers and industry teams can bring their own tasks, develop policies in simulation, and deploy them on physical robots.
I want a world full of robots — and I want more than 80% of them to use something I built.
I lead an initiative to connect policies developed in simulation with evaluation on physical robots.
Grounded co-training raises success from 52% to 86% on a dynamic dexterous pick-and-sort task.
AI-powered agricultural robotics, from delicate harvesting to crop monitoring. Acquired by Industrial Next in 2026.
I earned my B.S. in Mathematics and Computer Science and M.Eng. in Computer Science and Engineering at MIT, my M.S. in Robotics at Carnegie Mellon University, and my PhD in AI & Robotics at the University of Washington.
I was part of MIT’s DARPA Robotics Challenge team and coordinated the University of Washington’s SARA team on autonomous off-road navigation. Patented technology from SARA is now at Overland AI (autonomous ground vehicles for defense).