Anduril’s Brian Schimpf Spent Nine Years Warning of a Munitions Crisis. The Iran War Proved Him Right.

Brian Schimpf handed a reporter a three-pound tungsten warhead at a secretive West Texas test site, then watched an Anduril Bolt drone dive at 85 degrees toward hills the company would not let a photographer shoot. That scene opens a long Fortune profile of the 42-year-old engineer who runs America’s most-watched defense startup, and it lands at a useful moment. Anduril raised $5 billion in mid-May at a $61 billion valuation, roughly double its mark from June 2025. Its air defense systems were deployed during the war in Iran. And the munitions shortage that war exposed is the exact problem Schimpf has spent nine years telling the Pentagon it had.

The warhead is built for the Bolt-M, the munitions-carrying variant the U.S. Marine Corps put under a $23.9 million contract in January for more than 600 systems. I went through the Bolt-M’s specs when I covered that award earlier this year, and the detail that stuck with me is the one Schimpf keeps circling back to in the Fortune piece: the math. “Every bit of math matters,” he says. “You have a physics problem.” Three pounds of explosive, or four, against the aerodynamics of a backpack-sized airframe. That tradeoff is the entire product.

Schimpf is the engineer-CEO, not the founder you have heard of

Anduril’s public face is Palmer Luckey, the Oculus creator whose 2017 firing from Facebook preceded the company’s launch. Schimpf is the one running it, and has been since inception. He joined Palantir straight out of Cornell and left a decade later, as director of engineering, to help build Anduril’s first product. He describes himself as a “minority-opinion Democrat” who fundraised for Joe Biden, which complicates the tidy story of Anduril as a Trump-aligned enterprise.

That first product was Sentry, a line of autonomous solar-powered surveillance towers Schimpf built while living out of a trailer in the California desert in 2017. The September 2017 demo worked, the towers became Anduril’s first sale, and the data layer underneath them grew into Lattice, the software platform the company now sells as its core. The U.S. Army just consolidated more than 120 separate Anduril procurement actions into a single enterprise agreement worth up to $20 billion over ten years, built around Lattice. There is no money attached on day one. What it changes is how fast the Army can buy from a single vendor.

The Iran war turned Anduril’s thesis into a stockpile crisis

Anduril’s founding argument is that the U.S. cannot build weapons fast enough or cheaply enough for modern war, and that legacy contractors are too slow to fix it. Operation Epic Fury made that abstract case concrete. The 38-day bombing campaign against Iran, which ran from late February until an April 7 ceasefire, drew down American munitions at a rate analysts called unsustainable.

“We’ve depleted something like 30% of the Tomahawk stock,” Schimpf tells Fortune. “We shot a decade of production in about three days.” The Center for Strategic and International Studies put the figure higher, estimating more than 1,000 Tomahawks launched against an average annual procurement of 86, with replenishment stretching to 2030 or 2031. The same analysis found roughly half the THAAD interceptor inventory and nearly half the Patriot stockpile expended. “Our munitions production rate and our stockpiles are at a dangerous level,” Schimpf says. “War is inherently industrial in the modern era, and we are not prepared for that.”

The drone side of that war reinforced the point from a different angle. The U.S. Air Force lost 24 MQ-9 Reaper drones over Iran, roughly $720 million in airframes, because a $30 million aircraft built for permissive airspace cannot survive an integrated air defense network. That attrition is the strongest argument anyone has made for cheap, attritable, autonomous platforms, which is precisely the category Anduril builds.

Anduril beat the primes on the Air Force’s wingman drone

In April 2024 the Air Force narrowed its Collaborative Combat Aircraft program to two companies, Anduril and General Atomics, eliminating Boeing, Lockheed Martin, and Northrop Grumman from that phase. Anduril’s entry is the Fury, now designated YFQ-44A, which completed its first flight in October 2025 and flew semi-autonomously from the start with no operator on a stick. Production began at the company’s nearly $1 billion Arsenal-1 factory in Pickaway County, Ohio, roughly four months ahead of the original July 2026 target.

The broader portfolio Schimpf described to Fortune is wide. Lattice is positioned to anchor counter-drone programs across the U.S. military. The Ghost Shark autonomous submarine is under long-term contract with the Royal Australian Navy. The company sells to Australia, Japan, Taiwan, the U.K., and South Korea. Schimpf’s growth plan leans on selling the Altius loitering munition in volume to Taiwan and Ukraine, and on the Barracuda missile finding traction as a long-range strike weapon. Anduril confirmed to Fortune that it projects $4.3 billion in 2026 revenue, up from $2.2 billion in 2025, and is currently far from profitable.

The Altius record in Ukraine is the part the profile handles carefully

Schimpf does not pretend the hardware has been flawless, and Fortune does not let the question slide. One of the most public criticisms of Anduril is that the Altius drone it sent in large numbers to Ukraine did not work well, with reports of crashes, missed targets, and poor performance against Russian electronic warfare. A Reuters investigation late last year, which I covered in detail, documented two Altius drones nosediving during Air Force tests at Eglin, a Ghost X spinning out during a U.S. Army exercise in Germany, and Ukrainian forces largely setting the systems aside. Anduril’s own blog later acknowledged early effective hit rates of only 10 to 15 percent under intense jamming.

Schimpf’s framing to Fortune is that Altius was designed for launch from aircraft, because U.S. forces usually fight far from their bases, and that Ukraine needed something cheap, land-based, and mass-producible instead. He points to the F-35’s long delays and to aircraft carriers that run well over $10 billion each as evidence that hard defense hardware fails everywhere, not just at startups. “This stuff is the hardest of the hard,” he says. “It takes some iteration, and even then, it may work for a while until it doesn’t.” That is a candid answer. It is also the answer a company gives when its flagship export drew sustained criticism in the field. Stanford defense innovation expert Steve Blank, quoted in the profile, argues the real threat to Anduril is political rather than technical: “Congress is coin-operated,” he says. “If they fail it’s because they got out-lobbied and outplayed.” The primes Anduril is chasing have had lobbyists working Capitol Hill longer than Schimpf has been alive.

DroneXL’s Take

Two things about Anduril have been true at the same time for a while now, and most coverage picks one. The contracts are real, the production ramp is real, and the company has genuinely pushed its way into rooms that were closed to non-incumbents for decades. The field performance has also been uneven, and the Reuters reporting I wrote up last November was not a hit piece, it was a documented gap between marketing and battlefield results. The Fortune profile is notable because it holds both at once, and because Schimpf himself does not run from the second part.

What changed this spring is the backdrop. When I covered the Altius crashes, the easy read was that Silicon Valley had oversold itself to the Pentagon. Six months later, Operation Epic Fury burned through munitions faster than American industry can replace them, and 24 Reapers fell out of the sky over Iran. The argument Anduril has been making since 2017, that the U.S. needs cheap mass and fast iteration more than it needs exquisite platforms, stopped being a sales pitch and started looking like a description of the problem. That does not retroactively make Altius good. It does make the thesis harder to wave away.

The unresolved question the profile raises and does not answer is whether iteration speed actually closes the gap against a peer adversary, or only against Iran. Schimpf says time will tell what Iran means for Anduril’s tech, and that if a future conflict protracts, the Pentagon will test new weapons out of necessity. He is right that Iran is not China or Russia. The CSIS report said as much: the U.S. has enough munitions for any plausible Iran scenario, but the drawdown opened a window of vulnerability for a Western Pacific fight. Whether Anduril’s factories in Ohio and Long Beach can produce at a rate that matters before that window matters is the thing to watch, and it is not something a $61 billion valuation answers. Watch the CCA production decision, expected this fiscal year, for the first hard signal on whether the Fury moves from prototype to fielded aircraft. That is a real scheduled process with a real outcome, not a guess.

Source: Fortune.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.