YC F26 startup Preload captures synchronized video, EMG and tactile data for robots
ycombinator · x · 2026-10-02
Preload (YC F26) has released a launch video showcasing its end-to-end robotics data collection service: a head-mounted camera (egocentric vision), a sensor glove (pressure across fingers and palm), and an EMG band on the forearm — three signals recorded in sync, capturing how hard human hands hold, press, and let go.
Their thesis: video flattens the physical world into a projection, and force/contact/motion information lost at capture can't be recovered by better model architectures — so force must be measured at the data source.
The pipeline runs collect → synchronize/filter/annotate → train and evaluate against ground truth → deploy on real hardware. Target customers: frontier AI labs needing force-aware training data for VLAs and world models, humanoid/dexterous-hand robotics teams, automotive manufacturers with skilled manual assembly, and industrial automation inside existing facilities. The team comes from BCI companies decoding neural signals, with the full stack — collection, evaluation, policy training, real-hardware testing — built in-house. They claim collection at scale can start as soon as tomorrow.
More from Embodied
- Hugging Face LeRobot adds humanoid support: π0.5 policy drives 29-DoF Unitree G1 — freelerobot · 2026-10-02
- Researcher's blunt verdict on humanoids: "The robots do not work" — andrew_n_carr · 2026-10-02
- Robot's Early Taichi Policy Trained on Kaggle Shows Promising Kung Fu Moves — freelerobot · 2026-10-02
- BeyondSCe grasps objects by past-event reference, hitting 77% success on occluded targets — SeoulNationalUniv · 2026-10-02
- NUS's APPL uses structural priors to bridge skill learning and composition in robot manipulation — NationalUniversityofSingapore · 2026-10-02
- Open-source Harness fork moves coding agents out of the app into orca — dee_hw · 2026-10-02