The data layer for physical AGI.
We capture the whole-body, contact-rich, force-labeled movement that video can't see.
AI left the screen.
For a decade, AI lived in text and pixels. The frontier has moved into the physical world — and the industry is converging on the same conclusion: the next defining AI companies will be built in atoms, not just bits.
WHOLE-BODY CAPTURE
Video can't feel.
Every frame of internet video shows what a body did. None of it records what the body felt — the grip, the load, the resistance. That signal only exists at the moment of contact, and it has to be captured at the source.
PRESSURE-SENSING GLOVE
Now it can.
A sensing mesh across the palm and every finger reads grip, pressure, and load at the instant of contact — the force signal, captured exactly where the hand meets the world.
FINGERTIP TO FULL BODY
The whole body, in sync.
Modular nodes trace motion and force from each fingertip up through the arms, torso, and legs — one synchronized recording of everything the body does and feels. This is the data layer for physical AGI.
THE ON-BODY CORE
It all comes together here.
The core time-syncs every node — motion, grip, load, and first-person vision — into one continuous, force-labeled stream. Capture-ready data, straight off the body.
GROUND CONTACT
All the way to the ground.
Leg nodes and an instrumented sole capture gait, balance, and the ground reaction force of every step — closing the loop from fingertip to footfall.