The New Defense Post #45

Welcome back to the #45 edition of The New Defense Post!

In this edition, we’ll cover:

The Acutronic Group serves as the hidden backbone to much of the aerospace & defense industry, delivering the picks and shovels needed for next-generation systems.

We are the only manufacturer of micro-turbines, actuators, power systems, rate tables, and vibration test systems with a global footprint across the US and Europe.

UK-based Agon just emerged from stealth on July 29 with $30 million, which includes a $7M pre-seed from Lakestar, 201 Ventures and D3, followed by a $23M seed from XYZ, Lux Capital and Northzone, to build "synthetic" battlefield environments where AI-powered autonomous weapons train before they are deployed. (Tech.eu)

Founded by former Anduril director Tristam Constant and Cambridge Aerospace co-founder Junaid Hussain, with staff pulled from Applied Intuition, Google, Helsing, Palantir and Unity, Agon says its simulations let counter-drone systems practice downing swarms and let other autonomy platforms iterate faster than adversaries can adapt. (Tech.eu)

🗣 Tristam Constant, Agon co-founder: "We are in an AI arms race." He also said Agon is building the infrastructure to help European defence systems iterate and adapt at the pace of the threats they're meant to counter. (Tech.eu)

📰 Our Take: This is a picks-and-shovels play on the AI-in-warfare boom: instead of building autonomous weapons itself, Agon sells the training ground other companies' weapons practice in, so it profits from the competition regardless of which companies win contracts. That's a smart place to sit, and a founding team pulled from such leading tech giants and top-tier startups suggests technical credibility.

The catch is the one that's dogged military simulation for quite some years: the sim-to-real gap. A counter-drone system that flawlessly downs swarms in Agon's synthetic battlefield still has to prove itself against real jamming, real weather and an adversary who knows it was trained in a box. Worth watching whether "trained in Agon's environment" becomes a credential defence ministries actually ask for.

On August 10, UK air-defense startup Cambridge Aerospace closed a $300 million Series C,  led by new investor DFJ Growth (backer of Anduril, SpaceX and Tesla), at a $3.4 billion valuation with returning support from Lux, Accel and Lakestar. The round follows a $200 million Series B just four months earlier, a 2.6x valuation jump, and brings the company's total funding since its 2024 founding to roughly $636 million. (TFN)

Its Skyhammer interceptor, built to down Shahed-type attack drones for a fraction of the cost of the missiles usually fired to stop them, is already under contract with the UK MoD. A small interesting sidenote: Co-founder Junaid Hussain also co-founded Agon (issue's Spotlight #2), putting the same person behind two very different bets on how Europe fights drones. (TFN)

🗣 Wes Streeting, UK Defence Secretary: the Skyhammer contract is "exactly what our unicorn scheme is designed to create." (TFN)

📰 Our Take: Watch how small the pool of people making these bets actually is. Junaid Hussain co-founded Cambridge Aerospace in 2024 and Agon in 2026: two companies now worth a combined $3.4B-plus, built on opposite sides of the same problem: one intercepts drones, the other trains AI to fight them virtually. And Lakestar backed both. That's less a coincidence than a sign of how concentrated the UK's defence-tech talent and capital currently are.

The more interesting number is the one Streeting is happy to cite publicly: a real MoD contract instead of yet another paper valuation. But let us look at the competition: Stark at €3.5B-plus, Helsing at $18B, Quantum Systems at $8B, and now Cambridge Aerospace at $3.4B, all chasing the same finite pool of European defense budgets. All companies push onto the same market at once, and consolidation is on its way.

On July 29, the US Army awarded Lockheed Martin up to $58.6B to triple production of PAC-3 MSE Patriot interceptors to 2,000 a year by 2030, and therefore converted a one-year deal into a seven-year framework running through 2032. Days later, on August 3, Northrop Grumman signed a further $3B in agreements to become a second rocket-motor supplier for that same interceptor — breaking a monopoly L3Harris had held — and to boost THAAD motor production at its San Diego plant. (Reuters, Breaking Defense)

Both deals are a response to the current stockpile crisis: the Center for Strategic and International Studies estimated in July that the US has fewer than 1,000 Patriot interceptors and fewer than 250 THAAD interceptors on hand, after the war in Iran burned through inventories faster than they could be replaced. These are inventories a rotating cast of NATO allies, including Germany, Greece, the Netherlands, Poland and Spain, also draw on. (Military Watch Magazine, Reuters)

🗣 Michael Duffey, US Undersecretary of Defense for Acquisition and Sustainment: framework agreements like this one are "vital to accelerating the tripling of PAC-3 and quadrupling of THAAD interceptor production." (Breaking Defense)

📰 Our Take: The dollar figures are the least interesting part of this story. $61.6B in a single week sounds like decisive action, but both deals build toward 2030, which means that the actual shortage CSIS is putting at under 1,000 Patriot interceptors right now doesn't get fixed this or next year. For the NATO members who fly Patriot batteries of their own, the number worth tracking is not the contract value but the delivery schedule which makes European governments question their overreliance on US capabilities anew.

Generally, multiyear framework agreements are the right tool for exactly this problem. They give suppliers the demand certainty to expand capacity ahead of the appropriations cycle, rather than year to year. And making Northrop a second source for PAC-3 rocket motors (breaking a monopoly L3Harris had held) is a genuine supply-chain fix. It's worth separating "we finally diversified and committed to real capacity" from "the shelves are full again": they are not the same announcement.

NSPA Creates Fast-Track Procurement Route for NATO Counter-Drone Systems (The Defense Post)

Monava Closes Funding Round as Demand for Passive Drone Detection Grows (Tech.eu)

NATO's SACEUR Welcomes New NSATU Commander in Wiesbaden (NATO.int)

Loitering munitions and FPV drones have driven something like 70 to 80 percent of battlefield casualties and equipment losses in Ukraine since 2024, by one open-source estimate. This is the backdrop that makes the European Defence Agency's Sentinel Strike Challenge worth a closer look, despite prize money that wouldn't cover a single seed round. (Bruegel)

Launched in May to find Europe's best design for a 25-kilogram, 40-kilometer-range kamikaze drone, the challenge got  more than 140 entries. On July 29, EDA named five finalists: Germany's Helsing, the Netherlands' Destinus, Slovenia's C-Astral, Greece's Spirit Aeronautical Systems and Portugal's UAVision, plus five runners-up including France's Thales LAS, all of whom now advance to field trials at Portugal's Santa Margarida training area this autumn, competing for a shared €1.8 million prize pool. Actual winners won't be named until 2027. (The Defense Post)

What's notable is the mechanism besides the entrant list. Instead of one ministry anointing a national champion, EDA is running all ten finalists through the same live evaluation and collecting comparable performance data before anyone commits procurement money to a winner. That's a direct answer to a complaint EDA's own leadership has made publicly: chief executive Jiří Šedivý has described a European defense base fragmented by national purchasing preferences, producing small, localized markets instead of one large enough to support real competition, a well known critique. (Defense News)

Here's what that discipline costs though: time. Sentinel Strike opened in May, trials wrap in early October, and winners still won't be named until 2027. This is pushing toward two years to settle on a drone design, in a category where Ukrainian units are cycling new FPV variants in the field within weeks. A shared, comparative trial across ten competitors genuinely beats 27 ministries each buying their own favorite. It's just worth staying clear-eyed that rigor and speed are pulling in opposite directions here, and this format is built to optimize the former.

Love these insights? Forward this newsletter to a friend or two. They can subscribe to our website.