The India Decade

Mecka AI raises $60 million Series B led by Sequoia, with Nvidia and Microsoft’s M12 on board

The robot‑data startup secured a $60 million round to expand its human‑motion dataset platform, backed by Sequoia, Nvidia and Microsoft’s venture arm.

By The India Decade

robotics data lab with motion‑capture suits (file image)
robotics data lab with motion‑capture suits (file image) · “Foto Lab Robotics” by AraqiX (CC BY-SA 4.0) via Wikimedia Commons

Mecka AI clinches $60 million Series B funding

The San Francisco‑based startup announced Thursday that it has closed a $60 million Series B round. The financing was led by Sequoia, with Nvidia, Microsoft’s venture fund M12 and a handful of other backers joining the deal. The cash will be used to broaden the company’s sensor‑based data‑collection network and to accelerate the development of robot‑training models.

How the startup builds data for robots

Mecka AI’s core product is a platform that pays everyday people to perform routine tasks – from brewing a cup of coffee to changing a car tire – while they wear body‑mounted sensors and record the activity on their smartphones. The resulting motion‑capture data is then labelled and fed into machine‑learning pipelines that teach humanoid and industrial robots to mimic those motions. In the company’s words, it wants to be for robotics what firms like Scale AI have become for large language models: a reliable source of high‑quality, human‑generated training data.

Why the investors are interested

Sequoia’s lead‑investor status signals confidence that the robot‑data market is moving from niche labs to commercial scale. Nvidia, a hardware giant that supplies GPUs for AI workloads, sees a direct pipeline from Mecka’s datasets to the training of its own robotics‑focused AI stacks. Microsoft’s M12, the tech giant’s corporate venture arm, adds a software‑services perspective, hinting at potential integration with Azure’s robotics and AI services.

Industry context and competition

Mecka is not the first company to chase real‑world robot training data. XDOF, another data‑collection startup, has been reported to be raising a Series B at a $1.2 billion valuation. Larger data‑labeling firms such as Scale AI and newer entrants like Micro1, which began with language‑model data, are also expanding into the robotics arena. The crowded field suggests a rapid escalation of demand for high‑fidelity human‑motion datasets as manufacturers aim to ship robots that can operate safely alongside people.

What comes next for Mecka AI

With the new capital, Mecka plans to double the size of its contributor pool and to roll out more sophisticated sensor rigs that capture fine‑grained force and torque data. The company also hinted at upcoming partnerships with robot manufacturers, though no names were disclosed. If the $500 million valuation reported in earlier coverage holds, the funding round values the business at roughly that level, positioning it among the most well‑capitalised players in the nascent robot‑data ecosystem.

---

The round marks another milestone for the emerging market of robot‑training data, a niche that is quickly becoming a critical piece of the broader artificial‑intelligence supply chain.

Key numbers

WhatFigureSource
Funding amount$60 millionMecka AI announcement
Round typeSeries BMecka AI announcement
Lead investorSequoiaMecka AI announcement
Other investorsNvidia, Microsoft’s venture fund M12 and othersMecka AI announcement
Founded2024Mecka AI announcement
Valuation at round (reported)~$500 millionTechCrunch prior reporting

In this story

NameRole in the story
Mecka AIRobot‑data startup that collects human‑motion data for training robots

Topics

The day’s reporting, once each morning. No advertising.

More on this story