Agility Robotics co-founder and CEO Damion Shelton explains how warehouse work can lead to humanoid robots in homes, at a Stanford event with an audience of robotics and AI researchers.
Shelton says warehouses offer a practical first use for humanoids: moving totes in narrow aisles, lifting up to 25 kilograms, reaching from the floor to above six feet, and working at least 90 minutes between charges. He expects robots to move from logistics and manufacturing into less structured settings over time. Homes require much higher safety and manipulation capabilities, so he sees them as a later step.
A human-like form can fit spaces and workflows already designed for people, but each feature has a practical reason. Dynamic balance helps a narrow robot lift heavy loads; legs offer sideways movement and terrain access, while an upright torso helps with balance, sensors, and starting or stopping. Arms support manipulation, recovery from falls, and control of the robot’s motion. Shelton says the first commercial machines will weigh over 100 kilograms, with most of the torso occupied by batteries.
Shelton separates language and vision systems that interpret goals from physical intelligence that manages forces, contact, balance, and joint motion. He argues that hardware must support compliant, force-sensitive control, while reinforcement learning in simulation can discover robot-specific behaviors. Demonstrations can help teach higher-level tasks, such as grasping a tote, and language models can translate requests into plans. He expects these capabilities to work as a coordinated system rather than a single model controlling everything.
Useful robots are heavy and strong enough to cause harm, and dynamic robots can fall. Shelton calls for documented risk assessments, third-party review, human detection, and an independent supervisory system that can slow or stop the robot. He describes deployment as a gradual progression from warehouses with barriers to back rooms and more public settings, supported by extensive testing and operating experience.
Shelton expects human interaction to expand gradually, with robots first pausing or slowing around trained workers. He sees shared autonomy and waypoint commands as useful during adoption, while saying teleoperation is not a substitute for learning control in simulation. He also argues that robot hardware and software need integrated design, and that legs make sense where reach, balance, and versatility matter, even though wheeled platforms may suit other tasks.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Jonathan Hurst, co-founder of Agility Robotics, discusses Digit’s warehouse work, product development, safety, and the company’s path toward going public with a TechCrunch host at Disrupt 2026.
Peggy Johnson, CEO of Agility Robotics, discusses Digit 5’s safety, capabilities, and path to wider use with John Koetsier on NEXT.
Jonathan Hurst, co-founder of Agility Robotics, discusses the Digit 5 humanoid robot, its workplace uses and safety approach with Greg Nibler on FOX 12 Now in Oregon.
Jonathan Hurst, chief robot officer and co-founder at Agility Robotics, explains the software stack for humanoid robots operating in human environments.
Pras Velagapudi, CTO of Agility Robotics, explains how the company approaches data security across its robots, cloud services, and customer deployments.
Ankur Saxena, investment director at TDK Ventures, discusses humanoid robotics and physical AI with John Koetsier on NEXT with John Koetsier.
