YC S26. Tactile egocentric datasets recorded with its own synced capture hardware.



Pays businesses for recordings of their crews doing real work on a phone app, and supplies that footage to robotics labs.



Operates a network of human pilots and its own robot arms on real floors to produce manipulation datasets for embodied AI.
Markerless 3D capture of people at work from ordinary cameras or drones, retargeted to a robot and validated in physics.



Deploys robots into real-world sites for robotics companies and generates in-the-wild training data from those deployments.
YC P26. Records work at commercial sites through a large operator network for egocentric data, and trains operators to run robots on site.



Data engine for physical AI: egocentric video, motion, touch and voice data with multi-stage QC across 20+ countries.
Egocentric human-motion video from homes and factories, captured on head-mounted iPhone rigs through a contributor network run from Asia.
Pays working professionals per accepted hour to record their jobs on supplied cameras, from bars to repair shops, as data for robotics.



Builds large-scale egocentric multimodal datasets for robotics, wearable AI and embodied intelligence, with its own contributor network.
Egocentric human-action data captured with proprietary rigs and a factory network.
Network of operators capturing teleoperation and egocentric demonstrations, sold as robotics training data.



Buys camera recordings of people doing hands-on work, from home cleaning to skilled trades, as training data for physical AI and robots.



Data collection and labelling across the Middle East and Central Asia, specialising in egocentric data capture.
Optical motion-capture studio selling human motion datasets for humanoids, including the open BONES-SEED set.



Collects egocentric, UMI, teleoperation and tactile data; its ARC product monitors deployed robots and steps in live over VR when they fail.
The Robotics Map
A map of the robotics industry, from data and simulation to models, robots and deployments.
Filter Data producers
How to read this map ⌄
- Each station is one layer of the robotics stack, and its number counts the companies in it. The tiers read bottom-up, from data sources to what runs in the field.
- A line joins two layers that share companies, ones selling a real product in both. Its number says how many, and thicker lines mean more. Hover a line to see who; click it to keep the list open.
- +Zoom into a station, or click it, to open it. Companies only in that layer ring the station; companies that also work elsewhere sit on the line to that other layer.
- A company in four or more layers (today only NVIDIA) draws no lines. It sits inside every layer it works in instead.
Layers⌄
Busiest connections⌄
292 companies · 48 span more than one layer. Updated October 2026.
Maintained by Sam Padilla and Varun Nair