Mecka AI raises $60M Series B for humanoid robot training data

Mecka AI has closed a $60 million Series B round led by Sequoia Capital, with participation from Nvidia, TechCrunch reported. The startup pays individuals to record themselves performing everyday tasks—opening doors, folding laundry, stacking shelves—and then processes the footage into training data for humanoid and other robots.

The raise appears to build on earlier momentum in the space, with robotics startups pushing AI into physical environments, including Skild AI’s airport pilot and a $231 million spin-out from Mistral’s robotics team. Mecka’s approach mirrors broader trends in AI training data, such as Snorkel AI’s data-as-a-service model, which recently tripled its valuation to $3.5 billion.

What sets Mecka apart is its focus on human motion data. While most robotics startups train models in simulation or controlled lab settings, Mecka sources clips from hundreds of real-world environments. The company claims this reduces the “sim-to-real” gap that has delayed commercial deployment of humanoid robots.

The funding arrives as Asian startups begin scaling robotics beyond research labs. Mecka’s bet is that motion data will become a bottleneck as hardware improves, positioning the startup as a critical supplier to both robot makers and industrial customers. Whether that demand materializes beyond early pilots remains an open question.

Sources: techcrunch.com