Imperial spinout Machenta raises $40M to teach robots a new task from a single demonstration
Machenta, founded by Imperial College robot-learning professor Ed Johns, closed a $40M seed led by Creandum and Union Square Ventures. Its foundation model is pre-trained only on simulated data and then picks up tasks like packing or tidying from one human demonstration, across different robot bodies including humanoids. The money funds scaling the model and a new London lab.
Why it matters
Simulation-only pretraining could cut the costly real-world data collection that slows robot deployment.