Noble Machines Exits Stealth with Moby Humanoid, Already Deployed at Fortune 500 Customer

Noble Machines, a Sunnyvale startup founded by former engineers from Apple, SpaceX, NASA, and Caltech, has emerged from stealth mode with a humanoid robot already deployed at a Fortune Global 500 customer. The robot, called Moby, can lift up to 60 pounds (27 kg) and navigate steep inclines and outdoor environments, putting it in direct competition with humanoids from Figure, Boston Dynamics, and Agility Robotics.

The company, previously known as Under Control Robotics, went from founding to customer deployment in just 18 months. That timeline is aggressive even by Silicon Valley standards, and it signals a shift in how quickly new entrants can reach commercial readiness in the humanoid robotics space.

Illustration

Moby's hands are designed for heavy industrial manipulation tasks. Source: Noble Machines

The Noble Machines Story

Noble Machines was founded with a specific thesis: the humanoid robotics industry has spent too long on demos and not enough time on deployment. CEO Wei Ding and the founding team pulled together talent from organizations that build hardware under extreme constraints. Apple brought consumer product discipline. SpaceX contributed rapid iteration culture. NASA and Caltech added deep expertise in autonomy and control systems.

The company operated under the name Under Control Robotics during its stealth phase, quietly building Moby and lining up its first commercial customer. By keeping a low profile, the team avoided the hype cycle that has consumed several humanoid startups, choosing instead to let deployment results speak for themselves.

"Our customers are rethinking their operations in this new era enabled by AI and general-purpose robots that can collaborate with people," Ding said in the company's announcement. "We are accelerating this transformation, making industries safer, more efficient, and more resilient."

Under the Hood: Moby's AI-Driven Control System

Moby's standout technical feature is what Noble Machines calls "whole-body AI control." Unlike systems that treat locomotion and manipulation as separate problems, Moby uses a unified control architecture that coordinates its entire body for stability during complex tasks. When the robot lifts a 60-pound load, its legs, torso, and arms all adjust in real time to maintain balance.

The company claims end-to-end autonomy that lets Moby learn new real-world skills "in hours rather than months." This is a bold claim, and the specifics remain thin. What Noble Machines has disclosed is a multi-modal learning pipeline: Moby can be trained through language-based instructions, physical demonstrations, and gestures. In practice, this means a factory floor supervisor could show the robot a task once, describe the steps verbally, and have Moby replicate the workflow without writing a single line of code.

Moby also features adaptive mobility for outdoor and uneven terrain, including steep inclines. This sets it apart from warehouse-focused humanoids that assume flat floors. Construction sites, energy installations, and semiconductor fabs all present environments where ground conditions vary, and Noble Machines appears to have designed for these conditions from day one.

Who's Building Humanoids for Heavy Industry

Noble Machines enters a field that has grown crowded over the past two years. But the company's positioning is distinct: while most humanoid startups target warehouse logistics or light manufacturing, Moby is built for the physically demanding end of the spectrum.

Noble Machines differentiates on three fronts: payload capacity that exceeds most competitors, outdoor and rough-terrain capability, and a multi-modal training system that doesn't require robotics expertise to operate. Whether these advantages hold up as the company scales remains the open question.

What Moby Means for Construction, Energy, and Manufacturing

The industries Noble Machines is targeting (construction, energy, semiconductors, manufacturing, and logistics) share a common problem: labor shortages in physically demanding roles. The U.S. construction industry alone reported over 400,000 unfilled positions in 2025, according to the Bureau of Labor Statistics. Energy sector maintenance, semiconductor fab operations, and heavy manufacturing all face similar gaps.

Illustration

Noble Machines envisions Moby working alongside human crews in construction and industrial settings.

A humanoid that can lift 60 pounds, handle inclines, and learn new tasks from verbal instruction could slot into these roles without requiring facility redesign. That's the pitch, at least. The reality of deploying robots in unstructured outdoor environments is much harder than controlled factory floors. Dust, weather, uneven ground, and unpredictable human coworkers all add complexity that warehouse robots never face.

What's Coming Next

Noble Machines said it is preparing to release its next-generation robot "soon," though no timeline was given. The company will showcase Moby at NVIDIA GTC 2026 later this month, which will be most people's first chance to see the robot in person. GTC has become the de facto launchpad for physical AI companies, and Noble Machines will be sharing floor space with nearly every major humanoid startup.

The key milestones to watch over the next 12 months: how many units Noble Machines deploys beyond its initial Fortune 500 customer, whether the company discloses funding details (it has not announced any rounds publicly), and how the next-generation hardware compares to rapidly improving competitors. Figure AI, Boston Dynamics, and Agility Robotics are all shipping updated hardware in 2026, so the competitive window is narrow.

Frequently Asked Questions

The 12-Month Outlook

Noble Machines has done something rare in the humanoid robotics world: shipped a product to a real customer before most people knew the company existed. The 18-month timeline from founding to Fortune 500 deployment, if the claims hold up under scrutiny, would make Noble Machines one of the fastest humanoid startups to reach commercial deployment.

The next year will test whether Moby can scale beyond a single customer. Watch for funding announcements (the company will need capital to manufacture at volume), next-generation hardware details at GTC 2026, and head-to-head comparisons with Figure 03 and the new Atlas in real industrial settings.