New York-based Mecka AI has raised $60 million in Series A financing to scale its human-motion data platform for robotics training. Framework Ventures led the round, with Menlo Ventures, SV Angel, Kindred Ventures and Ted Xiao, a former Google DeepMind researcher and founding member of Project Prometheus, participating.
Capturing Human Motion For Robot Training
Mecka AI sells a stack that captures hand gestures, walking gaits and full-body kinematics from custom wearable sensors and consumer iPhones, then turns that data into training corpora used by robotics foundation models. The company says it has already collected thousands of gigabytes of human-demonstration data — the kind of physical-motion library that AI labs increasingly view as the limiting resource for embodied AI.
$100M Run Rate, 40 Employees
Mecka projects an annual run rate of $100 million based on signed contracts and currently has 40 employees. The $60 million Series A was raised in two tranches: a $25 million tranche closed in November 2025 and a $35 million follow-on. Co-founder and CEO Andrew Gao said the round will fund hardware capacity, R&D, and an expansion of the team that labels and curates physical motion data for robotics customers.
Robotics Data As The New Choke Point
The funding lands at a moment when robotics labs from industrial AI startups to large foundation model builders are scrambling for high-quality physical demonstration data. Without diverse, well-labelled human-motion corpora, even strong vision-language-action models struggle to generalise across robot embodiments and tasks. Mecka's pitch — that iPhones plus wearables can produce data orders of magnitude cheaper than teleoperated robot rigs — has resonated with a generation of physical-AI investors.
Founders
Gao and co-founder Mogen Cheng previously sold a payments-tech startup serving restaurants. Their third co-founder Jason Chong sold a company to Coinbase, and a fourth co-founder Duy Nguyen made his early money flipping sneakers. The founding team is unusual for the AI sector — none have credentials from frontier AI labs — but customers, including several physical-AI startups, have signed multi-year data contracts.
Reporting based on coverage from Fortune, BetaKit, citybiz and The SaaS News.