Most humanoid robots arrive with a highlight reel. Minerva Humanoids arrived with a job description and a customer. The San Francisco startup emerged from stealth this week with roughly $10 million in pre-seed funding led by General Catalyst, with Long Journey Ventures and Credo Ventures as co-leads. Its robot is called Roger. His job: go into places where humans shouldn’t.
The target list is specific: onshore and offshore oil and gas, explosive ordnance disposal, hazmat response. These are environments where the hazard is the whole point: toxic air, explosive atmospheres, crushing machinery. Minerva is honest about what autonomy can and can’t do today. Roger is semi-autonomous. Trained specialists operate him through a VR headset and force-feedback haptic gloves, staying physically outside the danger zone while the robot works inside it.
It’s a philosophy, not a compromise. Full autonomy in an unpredictable industrial disaster is a fantasy. A skilled human with robot hands in the room, via a robot body in the room, is a product. Minerva’s AI handles what machines are good at: balance, fall recovery, navigation, translating the operator’s commands into whole-body motion. The human handles judgment, improvisation, knowing which valve to turn.
Five months from concept to walking
That’s the number that stopped people. Five months to go from concept to a walking humanoid, in an industry where development cycles are measured in years. The team has the credentials to back it up. CEO Sandor Felber launched the humanoid research line at MIT’s Computer Science and AI Lab, then worked on Tesla’s Optimus program. CTO Maurice Rahme spent six years at Boston Dynamics, where he was the technical lead on Stretch from prototype to commercial product.
The supply chain tells its own story. The software and AI were developed in the U.S. The first Roger prototypes were manufactured in Germany. Commercial versions will be built in Germany and the U.S. with a Western supply chain and no single-source Chinese components. For energy and public-safety customers, that is a selling point.
The case for keeping humans in the loop
The robotics industry has a highlight-reel problem. Viral videos show humanoids doing backflips; the messy reality of useful work in ugly environments rarely makes the cut. Minerva’s answer is low-latency control and full-body operation that users learn in under five minutes. And a first paid industry pilot launching this month. Not a demo. A pilot. Paid.
The applications explain the urgency. Minerva estimates that around 7,000 energy workers die each year worldwide, the company’s own estimate, but the order of magnitude is the point. Every one of those deaths happened in a place a robot could have gone instead. Explosive ordnance disposal is the same story: the robot finds out what’s dangerous so the human doesn’t have to.
The company has already showcased Roger at the NATO Explosive Ordnance Disposal Centre of Excellence in Samorin, Slovakia, and is exploring deployments at two of the world’s ten busiest airports. Research partnerships with MBZUAI on energy capabilities and the University of Waterloo on perception in smoke, darkness and sensor failure cover exactly the conditions where humans currently risk their lives.
Not a backflip
For travelers, Roger’s first deployments will be invisible, and that’s the point. The robot that keeps an airport’s fuel farm safe or inspects an offshore platform is infrastructure you never see. For investors, $10 million pre-seed led by General Catalyst is a strong signal, and a paid pilot in month one of public existence is stronger. The risk is execution: commercial robots live or die on reliability, and reliability is earned in years, not months.
The humanoid race has a new entrant, and it’s the most grounded one yet. Not a backflip. Not a dance. A robot with a VR operator, haptic gloves, and a job in the worst room in the building.
Be First to Comment