Tracking the systems that turn robot demos into repeatable work.
Robotics Radar follows the full feedback loop: missing tasks, real-world data, hardware constraints, policy training, deployment failures, and the economics of keeping robots useful in the field.
Featured report Figure Index Turns Robot Data Into a Supply Chain Read the report →Written in English for a global robotics audience. Research notes, not investment advice.
Daily Robotics Radar
Daily notes, cleaned up into public briefs when they are worth keeping.
Robot Index
A structured reference map of humanoid OEMs, robot specs, deployment status, and supplier exposure.
Market Signals
A quick tape check for RoboStrategy, robotics ETFs, and physical AI proxies.
Company Files
Notes on exposure, evidence, risks, and second order effects by company.
The parts that decide whether a demo becomes labor.
Robotics Radar follows the messy middle: joints, hands, sensing, autonomy, safety cases, and the first deployments where the robot has to keep working after the camera leaves.
Torque, heat, and cost
The build only scales when motors, reducers, and thermal envelopes survive real duty cycles.
Hands before hype
End effectors turn demos into work. Grip, force control, and failure recovery are the tell.
Seeing the floor
Perception matters most in cluttered rooms, bad lighting, occlusions, and boring factory edge cases.
Hours in the field
The real score is uptime, support load, workflow fit, and whether customers ask for more units.
Where the robotics signal is reaching.
Public traffic view grouped into large regions. Counts update from live site visits; no personal visitor data is shown.
Counts are intentionally regional and aggregated: Asia, North America, Europe, and Rest of World. This keeps the signal useful without exposing personal visitor data.
Latest Radar
Figure Index Turns Robot Data Into a Supply Chain
Figure's Index contributor network suggests that the next robotics moat may be the speed at which a company identifies missing task coverage, acquires useful physical data, trains on it, and feeds deployment failures back into collection.
Callosum Raises $100M to Match AI Workloads Across Models and Chips
Callosum raised a $100 million seed led by Atomico to decompose AI workloads and route each task across different models and silicon under cost, energy, and latency constraints.
Veeda AI Raises $90M to Build Simulated Reality for Physical AI
Former NVIDIA Toronto AI lab leader Sanja Fidler launched Veeda AI with $90 million to build world models and simulated environments for training and evaluating Physical AI before deployment.