Asylon Robotics

A robot dog is walking a fence line in the dark. Nobody is cold, bored, or asleep.

It is 3 a.m. somewhere in America - a logistics yard, a stadium service road, a chemical plant - and the patrol is happening anyway. A four-legged robot trots the perimeter. A drone lifts off to check a sensor trip on the far side. Both report back to a single screen in a room in Pennsylvania, where a human is watching. This is Asylon Robotics, and this is roughly what it sells: the boring, dangerous middle of the night, automated.

Asylon builds US-made robotic security systems - autonomous drones, ground robots, and the software that runs them - and rents the whole thing out as a service. The company has logged more than 250,000 missions and put its robot dogs through over 150,000 miles of patrol. It is not a science project. It is a night watch.

Numbers per Asylon and 2025 funding coverage. The robot dogs, presumably, are not counting.

Guarding a big, empty place is expensive, repetitive, and occasionally dangerous - the three things people are worst at.

Physical security has a structural flaw: the work is mostly uneventful, until it isn't. Patrolling miles of fence line at a refinery or a distribution center means hours of nothing, punctuated by the rare moment that actually matters. Humans are costly to schedule around the clock, they get tired, and sending them into a possible gas leak or an unlit yard at 2 a.m. is its own liability.

Cameras help, but a camera only sees one spot and never walks toward the noise. The gap Asylon spotted sits between the static camera and the expensive human guard: something that can move, look, decide where to go next, and do it every night without complaint.

Three MIT roommates left aerospace to teach robots the night shift.

Damon Henry, Adam Mohamed, and Brent McLaughlin met as roommates at MIT and went on to aerospace careers - GE Aviation, Boeing, Johns Hopkins Applied Physics Laboratory. In 2015 they regrouped to start Asylon. Their bet was unfashionable at the time: that the hard part of security robotics wasn't the robot, it was making the whole system reliable enough that a Fortune 500 customer would trust it with a real site.

The proof point came in 2021, with a make-or-break demo for Ford - Asylon's first customer since 2019. The system, in Henry's telling, simply worked, all day, without drama. In a field full of impressive prototypes that fall over on day two, "it just worked" turned out to be the entire pitch.

Drones overhead, a dog on the ground, one brain watching both.

Asylon's system has four moving parts that behave like one. The clever bit is less any single robot than the fact that they coordinate - and that a human is always in the loop, verifying before anyone escalates.

It all ships as Robotic Security as a Service: roughly $100,000-$150,000 a year for the hardware, the software, and the people watching the screens. Customers don't buy robots. They rent a patrol.

From a dorm-room reunion to a $24M check

The customers are not early adopters looking for a gimmick. They run plants, stadiums, and supply chains.

Asylon's client list reads like a logistics-and-manufacturing roll call: Ford, Bayer, GXO Logistics, Kia, Mercedes-Benz Stadium, Home Depot, Moderna, Citizens Bank. These are organizations that measure security in dollars and incidents, not novelty. One Bayer deployment is cited at roughly $90,000 in annual savings per robot, with theft and vandalism reportedly dropping to zero on site.

The money agrees. The July 2025 Series B was led by Insight Partners, with strategic backing from Allegion Ventures (a physical-security insider), Veteran Ventures Capital, and Pennsylvania's GO PA Fund - a mix that points squarely at defense, logistics, and critical infrastructure.

Take the dull, dirty, and dangerous parts of security off people's plates.

Strip away the hardware and Asylon's mission is plain: make physical security more consistent and less hazardous by letting robots handle the parts humans do worst. The drones and dogs aren't there to replace the security team so much as to relocate it - out of the cold yard and into the operations center, where one person can oversee many sites at once.

That framing matters for a skeptical reader. The pitch isn't "robots instead of guards." It's "robots do the walking, people do the deciding." Whether that holds at scale is the open question every robotics-as-a-service company is being asked right now.

If autonomous patrol becomes normal, it probably looks a lot like this.

Asylon is not alone - Knightscope, Skydio, and a handful of others are circling the same future, and traditional guarding giants won't cede the night without a fight. What Asylon has is operating history: a quarter-million missions is the kind of number that's hard to fake and harder to catch up to. The Series B buys it time to widen that lead across defense and critical infrastructure.

The honest caveat: a service that costs six figures a year and depends on flawless reliability lives or dies on trust, and trust is slow. Asylon's whole strategy is to keep stacking uneventful, perfectly boring nights until "the robots have it" stops sounding strange.

So back to that fence line at 3 a.m. A year ago, securing it meant a person with a flashlight and a thermos, or a camera that could only watch one corner. Tonight the dog walks the line, the drone checks the far gate, and the only human in the loop is warm, awake, and watching twelve sites at once. The night didn't get less long. Asylon just stopped sending people out into it.