Brian Schimpf's Anduril wins Air Force production slot for autonomous fighter jets

Why it matters

Anduril's CCA win shows the Pentagon is willing to put a venture-backed defense company into combat-aircraft production, but not without preserving competition across hardware and autonomy software.

Brian Schimpf's Anduril won a U.S. Air Force production-related contract for its FQ-44 autonomous fighter jet on June 17, putting the Costa Mesa defense company inside one of the Pentagon's most consequential airpower programs and forcing its venture-backed manufacturing thesis into a federal procurement test.

The award, discussed by Schimpf on Bloomberg Technology, is part of the Air Force's Collaborative Combat Aircraft Increment 1 downselect. In the Air Force announcement, General Atomics' FQ-42 and Anduril's FQ-44 received engineering and manufacturing development and production contracts. The same announcement put Anduril into the mission-autonomy software pool, alongside General Atomics, Lockheed Martin, Northrop Grumman, RTX Collins Aerospace and Shield AI.

That second award matters because the Air Force is not buying these aircraft like a conventional fighter program. It is separating the airframe from the autonomy software, treating mission autonomy as a layer that is competed and fielded separately from the vehicle. For Anduril, the result is a two-front opening: FQ-44 is one of two aircraft selected for Increment 1 production, while Anduril's software team remains in the competition to provide the brains that will fly the broader CCA fleet.

Schimpf is not the louder founder in Anduril's origin story. Palmer Luckey brought the Oculus profile and the hardware mythology. Trae Stephens brought the Founders Fund and Palantir bridge. Schimpf brought the operating model: an engineer CEO who spent a decade at Palantir before joining Anduril in 2017, after earlier leading Cornell's autonomous-vehicle research program in the DARPA Grand Challenge and Urban Challenge. That history is newly relevant. The CCA award is the Pentagon asking whether the same software-first, autonomy-heavy approach can move from border towers and tactical systems into combat aircraft production.

A production win, not a clean sweep

The Air Force did not hand Anduril the whole program. General Atomics won the other hardware slot, and the autonomy competition remains open. The Air Force created a six-vendor software pool, with initial production options awarded to Anduril, RTX Collins Aerospace and Shield AI. The service plans to select a primary mission-autonomy provider for Increment 1 by summer 2027.

That structure keeps Anduril in a strong position without giving it lock-in. Anduril can build FQ-44 aircraft, but another vendor's software could still become the primary autonomy layer. Shield AI has already been closely tied to the aircraft: Anduril said in February that the YFQ-44A flew with mission-autonomy software from both Anduril and Shield AI. That is the point of the Air Force's architecture. The service wants portable autonomy, not a single vertically integrated contractor that owns both body and brain by default.

The contract value and exact number of FQ-44 aircraft covered by the award have not been disclosed. Breaking Defense reported that the Air Force is also withholding the number of vehicles each hardware provider will build in the first production lots. The service has said the broader goal is to procure more than 150 combat-capable CCAs by the end of the decade and approximately 1,000 over time.

The unanswered dollar figure is not a footnote. Anduril's promise to the Pentagon has always included a cost argument: autonomous systems that can create mass at a fraction of the price of crewed aircraft. The Air Force's own language around CCA emphasizes scale, affordability and continuous competition. Until unit counts and contract values are public, the scale of Anduril's immediate production win remains bounded by the program architecture rather than the headline.

The founder bet reaches manufacturing

Anduril's founders built the company against the idea that defense innovation had to move at prime-contractor speed. Schimpf, Luckey, Stephens, Matt Grimm and Joseph Chen started Anduril in 2017 with a thesis that software, autonomy and faster hardware iteration could win inside national-security procurement. For years, that thesis was easier to prove in systems adjacent to the core weapons platforms: surveillance towers, drones, command-and-control software, counter-drone systems and sensor networks.

FQ-44 is different. This is a front-line air-combat program designed to fly with crewed fighters, extending reach and survivability in contested environments. Anduril describes Fury, the aircraft tied to FQ-44, as an autonomous air vehicle built around Lattice software, modular payloads, commercial off-the-shelf jet engines and swappable autonomy software. The product choices match the founder thesis: avoid bespoke supply chains where possible, keep the software layer upgradeable, and design for manufacturability before the Pentagon asks for production.

That is also why the timing is notable. Anduril has spent the past year raising and deploying capital at a scale that looks less like a software startup and more like an industrial company. In May, TechCrunch reported that Anduril raised a $5 billion Series H at a $61 billion valuation, led by Thrive Capital and Andreessen Horowitz, after a prior $2.5 billion round at a $30.5 billion valuation. TechCrunch also reported Anduril doubled 2025 revenue to $2.2 billion.

The round was not just balance-sheet theater. Anduril has been building factory capacity to support the pitch it is now making to the Air Force. Its Arsenal-1 project in Ohio has been positioned as a more than 5 million-square-foot manufacturing site, and California officials in January highlighted an approximately $1 billion Anduril expansion in Southern California. The CCA award gives that manufacturing buildout a sharper purpose: produce combat aircraft at a pace and cost the Air Force says it needs.

Why the Air Force split the aircraft from the autonomy

The Air Force's most important procurement decision may not be picking Anduril and General Atomics. It may be refusing to let either company own the entire CCA stack.

By decoupling hardware from mission autonomy, the Air Force is trying to avoid a familiar defense trap: a platform that becomes hard to upgrade because the software, hardware and contractor incentives are fused together. A government-owned reference architecture is meant to let software from any vendor in the pool be integrated, updated and ported across different aircraft platforms.

For startups, that creates both opportunity and pressure. Shield AI can compete for the autonomy layer without owning the airframe. Anduril can win the airframe and still be forced to prove its software in a separate performance process. RTX Collins Aerospace can remain relevant in a software competition even without a hardware award. Lockheed Martin, Northrop Grumman and General Atomics stay in the software pool even where the hardware outcome narrowed.

That is a healthier market design for the Pentagon than simply anointing a new prime. It also cuts against the easiest Silicon Valley story about Anduril. The company is not replacing the traditional defense industry in one stroke. It is being inserted into a procurement structure that rewards speed and modularity while preserving government leverage.

A win that raises the execution bar

For Schimpf, the award validates a founder path that has become central to the defense-tech boom: recruit from software companies, pair that operating discipline with hardware talent, raise private capital before the government budget catches up, and show up to federal competitions with working systems rather than slide decks.

But the next test is not narrative. It is delivery. The Air Force's approach and timelines mean Anduril will have to stand up manufacturing, scale suppliers and integrate autonomy while the customer deliberately keeps other vendors in the loop.

Schimpf's Anduril has spent nine years arguing that defense companies should behave more like product companies. The Air Force has now given it a product-company problem at military scale: build the aircraft, keep the software competitive, meet the cost curve, and do it while the customer deliberately keeps other vendors in the loop.