Tutor Intelligence raises $75M from a16z to rent out robot workers by the hour

Tutor, which started out of MIT's Computer Science and Artificial Intelligence Laboratory, puts fleets of internet-connected robots into real production sites before they are fully autonomous[3]. Remote operators the company calls "tutors" step in when needed, and every finished task becomes training data for Tutor's robot foundation model, which in turn lets the robots take on more work[1]. a16z says it backed the company because it focuses on getting robots to finish specific jobs reliably for customer after customer, rather than on debates about data, models or hardware[2].

Cassie is a robotic arm on a four-wheeled base that takes packages off pallets and puts them in containers, or the reverse, at up to 18 items a minute, with a vacuum gripper that can lift more than 50 pounds[4]. Sonny uses two arms and depth cameras to pick items off shelves and pack them, and Tutor wants to extend it to assembly and equipment maintenance[4]. Today the robots handle goods such as breakfast cereal, snacks, cosmetics, home goods and toys in some of the largest US manufacturing and logistics networks[1].

Customers don't buy the machines. Tutor charges by the hour or by the pick, with no upfront purchase and no long-term contract, and handles deployment, maintenance and upgrades itself[1]. Its robots now work in almost four times as many facilities as a year ago, and it expects more than 1,000 robots under contract by the end of 2027, all built in the US[1]. Nearly a third of customers that received robots in the past 12 months were existing customers adding more, and some installs take as little as eight hours, according to a16z[2].

Much of the robotics funding wave has gone to humanoids and lab demos. Tutor's round backs a different approach: put practical robots to paid work now, keep a human in the loop, and let the deployments themselves generate the data that general-purpose robot models are short of.