Manifesto

Robots are moving from labs into the real world: warehouses, homes, farms, cities, but there's a bottleneck nobody talks about.

Software ships fast because every change is tested automatically, push code, run tests, and know what broke in seconds. Decades of tooling make this possible, from CI/CD and test suites to monitoring and observability.

Robots have none of this.

You train a policy, run it on a robot, and it fails. So you SSH in, pull the logs, download the video, and scrub through camera footage trying to figure out what went wrong, one run at a time, hundreds of runs per week, with no patterns detected and no memory between sessions. Every session starts from scratch.

This is the iteration loop for every robotics team on the planet, and it's slow.

The bottleneck isn't training models; GPUs are fast. It isn't testing either; you can always run the robot. The bottleneck is understanding what happened afterward, at scale, across thousands of runs, automatically.

We believe this is a solvable problem.

Camera is the dominant sensor modality across every robot form factor: arms, mobile robots, drones, humanoids. Sensor data varies but the challenge is the same: turn raw physical world data into understanding.

Vision-language models can now watch a robot and reason what happened. Sensor analysis can detect dynamics invisible in video: force anomalies, control instability, hardware degradation. Agentic systems can reason over both, at scale, to reveal the patterns buried in unstructured data that no human could sit through.

We're building the system that understands what robots do in the physical world. We analyze every robot run automatically and surface what's happening, so engineers can act on evidence instead of intuition.

Is it a data problem? A policy problem? A hardware problem? The engineer knows their system. We give them the clarity to decide fast.

We read every rollout: what the robot sensed, what the policy commanded, what actually happened. No engineer can watch hundreds of hours of multimodal data and hold the pattern in their head. An agentic system can. It synthesizes every rollout into structured knowledge that compounds across the full test history.

This is a hard problem. Understanding physical behavior from raw data, across robot types, sensor modalities, and environments, is unsolved. We're not afraid of that. Hard problems worth solving are the only ones that matter.

If we succeed, robots iterate as fast as software. And when robots iterate fast, they deploy everywhere and diffuse into every part of society.