Jason Ma is Co-Founder at Dyna Robotics, a robotics company building general-purpose robots powered by proprietary embodied AI foundation models that automate repetitive tasks for businesses.

Jason was a research scientist at Google DeepMind and NVIDIA AI, where he pioneered algorithms for training foundation models to teach robots new tasks. He studied at the University of Pennsylvania and was lead author on Eureka, a widely-read paper on training robots with human-like dexterity.

Jason's research has been incorporated into production-level software including NVIDIA Isaac Sim and PyTorch's TorchRL library. His work received recognition as NVIDIA's Top 10 Research Project of 2023 and multiple conference spotlight presentations.

Dyna Robotics raised $120M Series A led by Robostrategy, CRV, and First Round Capital, with participation from Salesforce Ventures, NVIDIA, Amazon, Samsung, and LG, valuing the company at over $600M. The company previously raised $23.5M in seed funding.

Dyna's DYNA-1 foundation model achieved 99% success rate running autonomously for 24 hours, folding over 900 napkins at production-grade quality. The robots are deployed at customer sites across hotels, restaurants, laundromats, and gyms running 16 hours daily.