Agility Robotics Showcases Progress in Digit’s Heavy-Lift Training

A LinkedIn post from Agility Robotics highlights ongoing work to train its bipedal robot Digit to perform a 65-pound deadlift. The post suggests this effort is being used as a benchmark to push both hardware limits and training methodologies, emphasizing the need for whole-body coordination and robust actuators and joints.

According to the post, Agility is using simulation to incorporate different objects into Digit’s training, modeling load distribution, grip forces, and shifts in the robot’s center of mass. This approach is presented as enabling policies that translate into dynamically balanced lifting in real-world conditions.

For investors, the update points to continued progress in developing commercially relevant manipulation capabilities beyond simple walking and navigation. If successfully productized, heavy-lift and material-handling skills could broaden Digit’s use cases in logistics, warehousing, and industrial workflows, potentially expanding Agility’s addressable market.

The focus on simulation-driven policy training also suggests ongoing investment in software and controls, not just mechanical hardware. This could support higher-margin recurring revenue opportunities over time, for example through updates or specialized task policies, and may help Agility defend its position in an increasingly competitive humanoid robotics segment.