Antioch Raises $8.5M Seed Round to Build Simulation Tools for Physical AI
Antioch, a New York-based startup, has secured an $8.5 million seed funding round led by A* and Category Ventures, valuing the company at $60 million. The startup aims to become the 'Cursor for physical AI' by developing high-fidelity simulation tools for robot developers. Currently, the robotics industry faces significant challenges due to the scarcity of real-world training data and the high costs associated with building physical testing environments. Antioch seeks to bridge the 'sim-to-real gap,' ensuring that virtual training environments are realistic enough for robots to operate reliably in the physical world. Founded by Harry Mellsop alongside former executives from Transpose, Google DeepMind, and Meta Reality Labs, the company allows engineers to create digital instances of hardware and connect them to simulated sensors. This approach enables scalable testing, reinforcement learning, and edge case analysis without the need for expensive physical infrastructure. By leveraging models from Nvidia and World Labs, Antioch provides domain-specific libraries to help smaller companies accelerate development. The funding will support the refinement of these simulations, addressing a critical bottleneck in the deployment of autonomous physical agents and offering a cost-effective alternative to traditional data collection methods.
Wire timeline
Antioch Raises $8.5M Seed Round to Build Simulation Tools for Physical AI
Antioch, a New York-based startup, has secured an $8.5 million seed funding round led by A* and Category Ventures, valuing the company at $60 million. The startup aims to become the 'Cursor for physical AI' by developing high-fidelity simulation tools for robot developers. Currently, the robotics industry faces significant challenges due to the scarcity of real-world training data and the high costs associated with building physical testing environments. Antioch seeks to bridge the 'sim-to-real gap,' ensuring that virtual training environments are realistic enough for robots to operate reliably in the physical world. Founded by Harry Mellsop alongside former executives from Transpose, Google DeepMind, and Meta Reality Labs, the company allows engineers to create digital instances of hardware and connect them to simulated sensors. This approach enables scalable testing, reinforcement learning, and edge case analysis without the need for expensive physical infrastructure. By leveraging models from Nvidia and World Labs, Antioch provides domain-specific libraries to help smaller companies accelerate development. The funding will support the refinement of these simulations, addressing a critical bottleneck in the deployment of autonomous physical agents and offering a cost-effective alternative to traditional data collection methods.
TechCrunch