Humanoid
Reimagine Robotics Ltd. has emerged from stealth with technology that lets robots learn directly from human workers, aiming to make automation accessible to non-specialists. The startup’s approach, which it calls ‘monkey-see, monkey-do,’ allows anyone to train a robot by demonstration and correction.
Founded in April 2025 by Jonathan Scholz and former colleagues Oleg Sushkov, Akhil Raju, and Misha Denil, Reimagine Robotics is headquartered in London and Sydney. Scholz previously founded and led Google DeepMind’s Applied Robotics team for seven years. The company’s vision is that robots should learn on the job, just like a new colleague.
“A useful robot should be able to learn from the person doing the work,” stated Jonathan Scholz, co-founder and CEO of Reimagine Robotics. “They should be able to show it a task, put it right when it makes a mistake, and move on to the next problem. That is what it means for a robot to learn on the job.”
Instead of requiring specialist programmers for every task change, Reimagine Robotics lets workers demonstrate tasks, watch the robot attempt them, and correct errors on the spot. The company emphasizes that people remain central to its model, positioning the robot as a tool to amplify human labor rather than replace it. Scholz noted that workers can identify bottlenecks, teach the robot, and apply that capability wherever needed, freeing them from repetitive physical tasks.
The company’s focus on human-in-the-loop learning could make robotics more adaptable in dynamic industrial and service environments, though specific deployment timelines and target markets were not disclosed.
Source: The Robot Report