Istari Digital wants to certify the airplane before it exists.

There is a version of the aerospace business that is mostly about aluminum, titanium, and the difficult physics of making heavy things fly. And then there is the version practitioners actually live in, which is mostly about proof. Before an aircraft flies for a military customer, someone has to demonstrate - with paperwork, tests, and a stack of engineering artifacts tall enough to be its own airworthiness hazard - that it will not fall out of the sky. That proof is expensive. It is slow. And, crucially, most of the underlying work already exists as digital models scattered across tools that do not talk to each other.

Istari Digital, a roughly 70-person company headquartered at 705 Cambridge Street in Cambridge, Massachusetts, is built on a simple and slightly maddening observation: the engineering answers are usually already in the computer. The models ran. The simulations converged. What's missing is a connected, auditable, trustworthy thread that ties them together into something a regulator - or a certifying authority - will actually accept.

Enter the matrix, literally

The company was founded in 2022 by Will Roper, who previously served as the U.S. Air Force's assistant secretary for acquisition, technology and logistics, and who has spent years evangelizing the "digital thread" and Industry 4.0. Chris Benson joined as co-founder and chief technology officer. In February 2023 Istari came out of stealth with a $13 million seed round, notably backed by former Google chief executive Eric Schmidt. Trade press, reaching for the obvious metaphor, described Istari as inviting the defense industry to "enter the Matrix."

The metaphor is glib but not wrong. Istari's platform is meant to be the connective tissue that lets models, simulations, requirements, and test data flow across tools, teams, suppliers, and - this is the hard part - classified environments, without the sensitive bits leaking. The pitch, printed plainly on the company's own site, is four words long: "Faster engineering. Safer outcomes."

The nine-in-nine problem

Ordinarily you would treat a two-year-old startup's traction claims with the skepticism they deserve. But Istari's early signal is unusually legible. The company reported closing nine government and commercial contracts within nine months of leaving stealth, and then did the thing startups almost never do voluntarily: it put new partners on a waitlist. When a company turns away demand, it usually means it has found a problem that lots of people have and very few have solved.

The marquee example of that problem is an aircraft called the X-56A. In late 2023 the Air Force awarded Istari a $19 million contract to pioneer a program named - with a straight face and a nod to the Wright brothers - "Flyer Øne." The goal is audacious in the way that only very technical goals can be: to create and flight-certify a digital twin of an aircraft before the physical article is built, so that a Military Flight Release could, in principle, be earned in the virtual world first.

Skunk Works signs on

In August 2024, Istari unveiled a partnership with Lockheed Martin's Skunk Works - the legendary shop behind the U-2 and SR-71 - to attempt exactly that on a modified X-56A. Over roughly two years, Istari, the Air Force Research Laboratory (AFRL), and Skunk Works built what they describe as a decentralized test network: unit tests plus industry-standard engineering models, connected through the Istari platform and fed into automated pipelines that check performance against requirements in real time.

Practically, that has meant digitizing the airworthiness process itself. The team built automated flight-validation tools and defined "digital continuation points" - a way of outlining a virtual flight envelope before a real one begins. In September 2025 Istari and AFRL expanded the partnership to fold in more sophisticated evaluation of flight and ground test data, tightening how digital models are measured against physical results. The unglamorous name for the math that makes any of this believable is VVUQ: verification, validation, and uncertainty quantification. It is the difference between a digital twin that looks good in a demo and one a certifying authority will sign.

By early 2026 the Department of the Air Force had awarded Istari a further $8.6 million to stand up "Industry Øne," an effort aimed at breaking down digital-engineering barriers across the broader Defense Industrial Base - the sprawling network of contractors whose incompatible tools are, arguably, the real adversary here. Istari's wager is that the same plumbing that could certify one X-plane could rewire how the entire base builds and proves complex systems.

None of this guarantees the X-56A actually earns its digital certificate; ambitious first-of-their-kind programs slip, and "on track" is a phrase that does a lot of work in press releases. But the direction is coherent, the customers are real, and the underlying complaint - that engineering knowledge dies in disconnected tools - is one almost every hardware engineer will recognize.