“Danny Drace believes that solving general-purpose robotics is still fundamentally a research problem, not yet an execution or scaling problem.”