Just months after exiting stealth, XDOF is negotiating a Series B round at a $1.2 billion valuation, driven by its massive $50M annualized revenue.

  • XDOF is in late-stage talks for a Series B funding round at a $1.2 billion valuation.
  • The startup's annualized revenue is reportedly approaching $50 million.
  • 8VC is expected to lead the potential new funding round.
  • XDOF provides critical data pipelines for frontier AI and robotics companies.

In a rapid ascent that has stunned the venture capital community, XDOF, a startup specializing in real-world teleoperation data for general-purpose robots, is in advanced discussions to raise a Series B round. The deal, which could value the company at approximately $1.2 billion, comes just three months after the company emerged from stealth mode.

While XDOF was not initially planning to raise capital so soon after its $70 million Series A in June, its explosive commercial trajectory changed the landscape. Sources indicate that the company's annualized revenue is nearing $50 million, prompting major investors like 8VC to approach the startup for a new round. The Series A round previously saw participation from heavyweights including Thrive Capital, Andreessen Horowitz, and Spark Capital.

Read Also:Realme 16x Launched in India Featuring Massive 7,000mAh Battery and Dimensity 6300 SoC

Why This Matters

BozokMedia analysis shows that XDOF is positioning itself as the essential infrastructure layer for the next wave of AI. While Large Language Models (LLMs) were trained on the vast expanse of the internet, physical robots face a massive bottleneck: the lack of large-scale, high-quality real-world datasets. XDOF is solving this by acting as an outsourced data-supply chain for the entire robotics industry.

XDOF is effectively becoming the Scale AI for the physical world, bridging the gap between digital intelligence and mechanical execution.

Historical Background: The company was founded in 2024 by UC Berkeley researchers Philipp Wu (CEO) and Fred Shentu (CTO). Their work began with GELLO, a low-cost teleoperation system designed to allow humans to remotely control robotic arms to generate training data. This research laid the groundwork for the sophisticated data pipelines XDOF offers today.

Currently, XDOF is collaborating with UC Berkeley’s AI Research lab to launch ABC, which is poised to be the largest collection of high-quality robot training data ever assembled. To achieve this, the company employs a global workforce of human collectors who wear sensors to record everyday physical tasks, providing the nuanced data required for robots to navigate the real world.

MetricLLM TrainingRobotics Training (XDOF)
Primary Data SourceInternet Text/CodeHuman Teleoperation/Sensors
Scaling MethodWeb ScrapingPhysical Data Collection
ComplexityHigh (Semantic)Extreme (Physical/Spatial)
Did You Know?: XDOF's founders used a low-cost system called GELLO to prove that high-quality robot training data could be generated through affordable human-operated hardware.

Frequently Asked Questions

1. What makes XDOF different from other AI companies?
Unlike LLMs that use text, XDOF focuses on physical movement data required to train robots for real-world tasks.

2. Who are XDOF's customers?
XDOF is already working with approximately 20 customers, including several frontier AI laboratories.