Mecka Raises $60M Series B to Scale Data and Robot Deployment – ​​Unite.AI

Mecka, a company that builds data infrastructure for learning robots, announced on October 7, 2026 that it has raised a $60 million Series B led by Sequoia Capital. New investors include NVIDIA, Qualcomm Ventures, Samsung and M12, Microsoft’s venture capital fund, the company said, with continued support from Kindred, Framework Ventures and Neo.

Participating angel investors include DoorDash founder and CEO Tony Xu, former ServiceNow and Snowflake CEO Frank Slootman, and former Tesla Optimus boss Milan Kovac, according to the company. Mecka said the capital is expanding its data infrastructure, deepening its in-house research lab and developing commercial robot deployment, work the company describes as the data and deployment layer for robotics.

The company’s thesis is that what robots lack is experience. No one has ever written how a hand grips, bends or pours, how hard it presses or when it lets go. “Motion, contact, force, and geometry are not on the Internet. You cannot download, license, or buy them,” the company wrote in its Series B announcement. Mecka describes this gap as the biggest bottleneck in robotics, calls robotics the biggest future productivity opportunity, and predicts data will become a trillion-dollar market.

On its website, the company describes a series of challenges between raw activity and usable training data. Real-world tasks are noisy, so models need large-scale edge cases and task-specific demonstrations to generalize. Raw video isn’t enough: teams need annotations, context, quality assurance, motion understanding, and evaluation workflows before real-world activity becomes useful training data. Finally, the lab is not the finish line: robotic systems must be tested against real-world workflows, environments and constraints before they can work in the field.

The capture and search stack

Mecka said that to digitize the physical world he had to build the entire stack: acquire quality hardware, reconstruction models and infrastructure. The company designs and manufactures its own multisensor hardware for each signal class, with sensor selection and synchronization driven by what its models need downstream. It operates a global manufacturing operation, with acquisition fleets recording human demonstrations in homes and commercial environments.

An in-house lab builds computer vision and multimodal models that transform raw, noisy reality into structured signals, covering motion tracking, 3D reconstruction and sensor alignment. The company reports state-of-the-art sub-centimeter hand pose accuracy on in-kind data.

The company’s website lists signal and action primitives, modules it describes as ways of capturing or interacting with physical reality: pixel understanding, depth and geometry, object tracking, segmentation, tactile force, sound waves, velocity and motion, point cloud, surface contour, and localization. Mecka also offers an iOS app to capture environment- and task-specific data, as well as web tools to browse, visualize, and query data, collaborate on datasets with teams, and run inference APIs.

The EgoVerse dataset and transfer study

Mecka said he tested his bet on human demonstration in the real world with researchers from Georgia Tech, Stanford, UC San Diego, ETH Zurich, MIT and Meta through EgoVerse, a human-to-robot transfer study replicated across labs, tasks and robots. The EgoVerse paper, first presented on April 8, 2026 and last revised on July 7, 2026, describes a collaborative platform that unifies data collection, processing and access in a shared framework, enabling contributions from individual researchers, academic laboratories and industrial partners.

The current version of the document includes 1,362 hours of human demonstrations in 80,000 episodes, covering 1,965 tasks, 240 scenes, and 2,087 unique demonstrators, with standardized formats, relevant annotations for manipulation, and tools for downstream learning. Mecka’s FAQ describes EgoVerse as its large-scale human interaction dataset, built from real-world first-person activity across tasks, spaces and environments.

In addition to the dataset, the authors conducted a large-scale human-to-robot transfer study, with experiments replicated across multiple labs, tasks, and robot embodiments under shared protocols. The paper reports that policy performance generally improves as human data increases, but that effective scaling depends on the alignment between human data and the robots’ learning goals. The list of authors includes Josh Gao and Jason Chong.

Customers, revenue and model of the integrator

Mecka said it supplies many major frontier robotics labs and numerous Mag 7 companies. The company said it surpassed $100 million in full-scale revenue in June 2026, in just a few months of operation, and expects revenue of $300 million by the end of the year.

The same stack powers the company’s commercial deployments. Mecka describes itself as a modern robotics integrator: Where a traditional integration is delivered once and remains fixed, its implementations acquire data on-site, are post-trained, and improve with each hour of operation, according to the company. For companies that want physical AI but don’t have a robotics team, Mecka says it offers hardware, data acquisition, post-training, integration and continuous operations. Its FAQ states that Mecka doesn’t build robots; instead it connects to the hardware, model and business ecosystems and owns the data, integration and evaluation layer between them, an arrangement the company calls the Mecka Loop.

Mecka said the funding allows him to create more data faster: more tools, deeper research and robots deployed where the work is. The announcement is signed by Josh, Jason and the Mecka team, and the company said it is hiring staff in research, hardware and operations.