Skild AI, the Pittsburgh-based robotics intelligence startup founded in 2023, has hit a $100 million annual recurring revenue run rate just 10 months after beginning commercial deployments. Skild’s client roster has grown from eight customers at the start of 2026 to more than 60, with hundreds of robots now running its proprietary AI software across manufacturing, logistics, inspection, security, and food preparation.
The tech behind the traction
Skild’s core product, branded as Skild Brain and built on its S1 model, takes an approach that sidesteps one of robotics’ most persistent headaches: the need to retrain AI models for every new task. Instead of requiring thousands of examples and extensive weight updates, S1 learns from a single human video demonstration using in-context learning.
The S1 model reportedly achieves a 96% success rate on tasks it has previously encountered. On entirely novel tasks, ones the system has never seen before, it hits 66%. The model achieves that second figure from watching just one demonstration video, performing at a level equivalent to having been trained on roughly 380 examples.
A funding machine with serious backers
