Robotics data startup XDOF eyes $1.2 billion Series B round
XDOF, a robot-training data provider, is in late-stage talks for a Series B financing valued around $1.2 billion.
XDOF, founded in 2024 by UC Berkeley researchers Philipp Wu and Fred Shentu, has moved from stealth mode to courting a Series B round that could value the company at roughly $1.2 billion, according to sources familiar with the deal. The financing would be led by venture firm 8VC, with the startup’s swift revenue growth—approaching $50 million on an annualized basis—prompting the outreach. XDOF’s business model centers on constructing data pipelines, collection hardware, and annotation services that enable AI labs and robotics firms to train general-purpose robots, a niche likened to Scale AI for the physical world.
Its flagship data set, dubbed one outlet, combines remote teleoperation of robotic arms with body-sensor recordings of everyday tasks. To scale this effort, XDOF intends to hire worldwide teams of remote teleoperators and egocentric data collectors. The company already serves about 20 customers, including several leading AI labs, and previously raised $70 million in a Series A round backed by Thrive Capital, Andreessen Horowitz, Lux and Spark Capital.
Why it matters
The funding could accelerate the creation of large-scale robot training data, a key bottleneck for wider adoption of general-purpose robotics.
In this story
