Mecka AI, a startup focused on gathering and analyzing human motion data to improve humanoid robot training, is nearing a new financing round led by Sequoia Capital that could value the company at around $500 million. This potential deal follows a $60 million raise just three months ago, led by Framework Ventures with participation from Menlo Ventures, SV Angel, and Kindred Ventures. The exact size and terms of the new round have not been disclosed and remain subject to change.
Founded in 2024 by four entrepreneurs, including Canadians Josh Gao and Mogen Cheng and U.S.-based Duy Nguyen, Mecka AI aims to overcome a significant obstacle in robotics development: the lack of physical-world data. The founders, who do not have robotics backgrounds, identified that capturing real-world human interactions is crucial for advancing general-purpose and humanoid robots.
The company, named after the concept of "mecha"—fictional giant robots controlled by humans—collects data by paying individuals to record themselves performing everyday tasks using body sensors and smartphones. This egocentric data collection method helps train robotic models more effectively.
Mecka AI projects reaching an annual revenue run rate of $100 million by the end of 2026. Although the startup has not publicly shared its client list, its data is reportedly used by various robotics companies and AI research labs. This approach complements other data collection techniques such as teleoperation.
Mecka AI operates in a competitive landscape that includes startups like XDOF, which is reportedly nearing a $1.2 billion valuation, and established human-data platforms like Scale AI and Micro1, all contributing to the growing ecosystem of real-world data for robot training.