Asimov
Real-world human movement data for humanoid robots
Y CombinatorWinter 2026ActiveIndustrialsManufacturing and Robotics
What they do
Problem
Humanoid robot developers lack large-scale, diverse real-world human movement data needed to train robots to operate reliably across varied environments.
Solution
Asimov collects and delivers real-world human movement datasets from households and businesses to train humanoid robot models.
Who it's for
Leading robotics labs and humanoid robot companies building and training robot behavior models.
Founders
Lyem Ningthou
Founder
Defense tech robotics researcher → building the data layer for humanoid robotics
Open roles 2
Company timeline
- Founded2026
- YC batchWinter 2026
- StatusActive
- Team size3 people
- LocationSan Francisco, CA, USA; Remote
Similar startups
Cortex AI
Large-scale real-world robot & human data for embodied AI
Intelligence Factory
Human Intelligence for Robots
shotwell.ai
Automated Scale AI for robotics data
Physical Intelligence
Physical Intelligence is building a foundation model for robotics, br…
Almond
The most capable dual-arm robot for physical AI
Vision Lab
Industrial data layer for robotics training