Superhuman
Spatial Reasoning.
Software-Only.
Any Robot.

Humans can navigate any environment with ease to get stuff done. It’s time AI can do the same.

Humans can easily navigate between indoor and outdoor places with just our eyes.

Our brain passively builds internal models of the outside world, which we can reuse years later to recognize where we are. We can read a 2D map we’ve never seen and use it to navigate a maze-like 3D indoor space. We believe zero-shot navigation for robots is a problem that software-alone can solve.
A detailed 3D city model showcasing buildings and green lines, representing infrastructure and urban layout.

Backed by top-tier investors who share our bold vision

We’re proud to partner up with some of the most active investors in AI & robotics.
Felicis logo
Inovia logo
Naval Ravikant logo
Caltech logoNaval Wilson Hill logo

We innovate at the frontiers of foundation model research, while deploying into real products.

We are a small team of AI research + product pioneers redefining what spatial reasoning means for AI. We build zero-shot, hardware-agnostic AI systems, publish at top AI venues, and partner with top academic labs.
Available roles

Partner with us to enable navigation for any environment and robot.
No hardware retrofitting required.

We currently partner with some of the world leaders in autonomy. Learn how we could solve positioning + navigation without retrofitting any hardware to the robot or the environment, through just an over-the-air software update.