The $60M Series B round was led by Sequoia and will fund the development of custom multi-sensor capture hardware and a computer vision lab. This lab is designed to convert raw, noisy footage into structured training signals, and the company is currently hiring engineers to scale both the hardware and software operations.
