Launch
Runtime: Robots that think in simulation (opens ycombinator.com)
A simulation-based robot policy that needs no fine-tuning or teleoperation
LaunchConfirmed by the company
Hemlock: The missing force signal in robotic training data
Context: Hemlock pairs egocentric video with force data from workers' muscle signals; it says adding it improved pi0.5's success 10x on force-sensitive tasks.
From YC Launches: We collect egocentric video data + force and contact dynamics derived from workers' muscular signals. Adding Hemlock's force data to human training demos improved pi0.5's success by 10x and Gemini Argon's by 2.7x on force-sensitive manipulation tasks.
- Source
- YC Launches (ycombinator.com)
- Published
- First seen here
- Edition
- Tuesday, October 27, 2026, ranked 2
- Status
- The company itself announced or confirmed this.
- Topics
- Y Combinator, Robotics, AI
