Our story
Robots are getting good at seeing and talking. Acting in the physical world is much harder. To learn how to handle real objects, they need to learn from people doing ordinary tasks, recorded from a first-person point of view. That kind of data is scarce, and most of it is collected outside Europe with little attention to privacy or consent.
We started EgoVista to build that missing layer in Europe. We collect everyday manipulation video from a network of contributors, anonymize it, and annotate it into datasets that robotics teams can train on directly. European by design, compliant by default, and built to the formats the field already uses.
What we do
We handle the full chain: collection on spec through our contributor network, GDPR-compliant anonymization, and multi-layer annotation, delivered as training-ready datasets (LeRobot and RLDS compatible). Teams that already collect their own footage can send it to us for annotation only. Data is collected and processed in the EU, and faces are anonymized before any external processing.
The team
Léonard Docquier
Co-founder & CEO
Léonard founded EgoVista and leads strategy, product, and customer work. He approaches robotics from the application side: how teams actually source and use training data, and what it takes to make European data collection trustworthy and compliant.
Howard Hao-Yu Hou
Co-founder & CTO
Howard leads engineering and the annotation pipeline. He holds a master's from the University of Toronto, is a first-author at ICCV, and specializes in 3D perception. He owns the work that turns raw video into clean, structured, training-ready datasets.
Alessio Palma
Co-founder & Chief Scientist
Alessio leads research. He holds a PhD in robotics focused on 3D perception and manipulation, with work published at ECCV, and a background spanning Sapienza Rome and TU Darmstadt. He shapes the science behind EgoVista's most advanced annotation layers.

EgoVista is a member of the NVIDIA Inception program, which supports startups building the next generation of AI and robotics.
© 2026 NVIDIA, the NVIDIA logo, and NVIDIA Inception are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries.