Mecka AI raises $60 million to train robots with human motion data from body sensors and iPhones

Mecka AI, a New York City–based startup, raised $60 million across two rounds: a $25 million Series A in November and a $35 million follow-on investment. Framework Ventures led both rounds; other backers included Menlo Ventures, SV Angel, Kindred Ventures, and angel investor Ted Xiao.

The company trains robotics models on data collected from humans—captured via custom body sensors and iPhones—rather than from teleoperation (manual remote control of robots). Gao said this approach was explored mainly in research papers but his team showed it could work at scale.

Mecka AI projects an annual run rate of $100 million based on already-signed contracts and currently has 40 employees. Cofounder and CEO Josh Gao stated that useful robots can be deployed today.