Human Archive is a data lab modeling human embodied intelligence.
We're researchers from Stanford, Berkeley, NYU GRAIL, UPenn GRASP, JHU Robotics, and national robotics champions, backed by Y Combinator and $8.2M.
Hardware: We design, manufacture, and calibrate thousands of custom collection rigs, gloves, and wrist cameras.
Operations: We have 100+ full-time team members collecting large-scale multimodal video and sensor data from 100,000+ contributors and 500+ industry partners across homes, hotels, restaurants, agriculture, manufacturing, construction, and retail in 30+ countries.
Software: We develop custom 3D perception and annotation models tailored to our hardware and train robot policies on our datasets. Our SLAM and hand-pose models achieve state-of-the-art performance.
Despite decades of research, we still barely understand how humans think, move, or interact with the world. Human Archive's datasets are foundational infrastructure for automating manual labor, increasing global abundance, and advancing our understanding of the brain, cognition, spatial computing, and prosthetics.
If you're a researcher, click here and we will send you our customer dashboard and samples. Currently OTS we have RGB, RGB-D, audio, tactile, mocap, and robot data.
Feel free to email raj@humanarchiverobai.com or want to chat? Request time here.