Mecka AI Approaches $500M Valuation as Sequoia Leads Fresh Funding Round

Mecka AI, which specializes in gathering and processing human motion information to power humanoid and other robotic systems, is on the verge of securing a Sequoia Capital-led funding round valuing the company at roughly $500 million, according to sources familiar with the transaction.

This latest investment arrives merely three months following Mecka's announcement of a $60 million funding round spearheaded by Framework Ventures, which also drew backing from Menlo Ventures, SV Angel, and Kindred Ventures.

The exact amount of the current round remains undisclosed to TechCrunch. Additionally, the deal structure has not been finalized and may still undergo modifications before closing.

Requests for statements from both Mecka AI and Sequoia Capital went unanswered and unacknowledged, respectively.

Mecka AI was established in 2024 by a quartet of founders: Josh Gao and Mogen Cheng, both from Canada who previously launched a restaurant technology company focused on financial services, and Jason Chong, who moved to Coinbase following the company's acquisition of his cryptocurrency exchange platform. Duy Nguyen rounds out the team as the operations lead and is the sole non-Canadian member.

Despite lacking prior experience in the robotics sector, the founding team identified a critical gap: the scarcity of real-world physical data and recognized that collecting authentic environmental interactions represented the main constraint preventing the development of general-purpose robotic systems, particularly humanoids.

Taking its name from "mecha," the term for human-piloted mechanical robots in fiction, Mecka aims to replicate the success that Scale AI, Mercor, Surge, and comparable human data firms have achieved in the large language model space. The company compensates individuals to document themselves executing routine activities—such as brewing beverages or performing vehicle repairs—through body-mounted sensors and mobile devices.

During early June, Mecka was forecasting it would reach a $100 million annual run rate by the conclusion of 2026, as Gao shared with Fortune at the time of the company's prior fundraising announcement.

Though Mecka AI has not made its roster of clients public, numerous robotics enterprises and artificial intelligence research organizations depend on real-world information gathered through this "egocentric" methodology, combined with additional physical data acquisition techniques including teleoperation, to develop their systems.

The competitive landscape includes other firms gathering real-world information for robot development, such as XDOF, which TechCrunch reported was approaching a funding round valuing it at $1.2 billion last week, alongside broader human-data platforms like Scale AI and Micro1 that are expanding their focus beyond language models.