Perseus Defense Is Building Cheap Guided Missiles to Kill Cheap Drones

The Macro: The Pentagon Has a Math Problem

I want to talk about the most embarrassing equation in modern warfare. A $500 commercial drone can destroy a $100 million asset. That is not a hypothetical. It has happened in Ukraine, in the Red Sea, and in conflict zones across the Middle East. Militaries around the world are watching cheap drones neutralize expensive equipment and they do not have a good answer.

The current counter-UAS solutions are almost comically mismatched to the threat. The US Navy has used SM-2 missiles that cost over $2 million per shot to take down Houthi drones that cost a few thousand dollars. Patriot interceptors run north of $4 million each. Even the cheaper options like Coyote interceptors from RTX (formerly Raytheon) cost around $100,000 per engagement. When your adversary can build ten replacement drones for the cost of one of your interceptors, you are losing the attrition math before the fight starts.

The Department of Defense knows this. The Replicator initiative, launched in 2023, was designed to accelerate autonomous and low-cost systems. DARPA has been funding counter-UAS research across multiple programs. The Army stood up Joint Counter-small Unmanned Aircraft Systems Office (JCO) specifically to coordinate the response. Every branch is scrambling.

But the defense primes have a structural problem. Lockheed Martin, Northrop Grumman, and RTX are optimized for expensive, exquisite systems with long development timelines and massive margins. They are not set up to build a $10,000 interceptor at scale. Their cost structures do not allow it. Their procurement relationships do not incentivize it. The companies best positioned to solve this problem are the ones least motivated to solve it.

That is the opening. The counter-UAS market is projected to exceed $10 billion by 2030 and the incumbents are leaving the bottom of the market wide open.

The Micro: NASA Engineers With a Two-Week Build Cycle

Perseus Defense is building man-portable, self-guided micro-missiles designed to intercept drones at a fraction of the cost of existing systems. The target price point is under $10,000 per engagement, which would represent roughly a 90% cost reduction compared to current solutions.

Jason Cornelius is the co-founder and CEO. Steve Messinger is the co-founder and CTO. Both come from aerospace backgrounds at NASA and Boeing, with experience in spacecraft design and autonomous systems. They are based in Austin, Texas and went through Y Combinator’s Summer 2025 batch.

What caught my attention is the development velocity. In eight weeks following a strict two-week design-build-test cycle, the team went through four major guided missile iterations with over 30 live-fire flight tests. That is an absurd pace for a hardware weapons system. Traditional defense contractors take years to reach that milestone count. Perseus is treating missile development like software sprints and it appears to be working.

The platform is described as multi-domain capable, meaning it can be deployed across land, sea, and air scenarios. The man-portable aspect is important because it means small units can carry and deploy the system without vehicle-mounted launchers or heavy logistics support. That maps directly to the way the drone threat actually manifests: distributed, mobile, and unpredictable.

Their customers are the Department of Defense and the Department of Homeland Security. DHS involvement is notable because it signals a domestic counter-drone mission beyond the battlefield. Think border security, critical infrastructure protection, and event security. The FAA has been dealing with unauthorized drone incursions at airports with essentially no tools. That is a real market beyond the military.

Perseus is competing against the interceptor programs from Anduril, which has the Anvil kinetic interceptor, and RTX’s Coyote. Anduril has significantly more funding and a broader defense platform, but their interceptor is one product among many. Perseus is focused entirely on this problem. Shield AI has drone-on-drone capabilities but is primarily a software and autonomy company. The pure-play focus on mass-manufactured guided missiles at rock-bottom costs is Perseus’s differentiator.

The Verdict

The counter-UAS market is one of those rare defense opportunities where the need is urgent, the incumbents are slow, and the physics favor the attacker. Perseus Defense is attacking the cost problem directly with a hardware-first approach and a development speed that looks more like a YC software startup than a defense contractor.

The risks are real. Defense procurement is a nightmare for startups. Getting from prototype to production contract involves security clearances, testing standards, and bureaucratic timelines that can eat a small company alive. Anduril has spent years and billions building the relationships and infrastructure needed to sell to the Pentagon. Perseus is two people in Austin.

In thirty days, I want to know if they have active contracts or letters of intent from DoD or DHS. Sixty days, the question is manufacturing scale. Can they produce these interceptors in volume at the price point they are targeting, or does the cost climb when you move from prototype to production? Ninety days, I want to see a competitive shoot-off result against Coyote or Anvil. The pitch is compelling. The founders have the right backgrounds. The market is desperate for this product. Whether a two-person team can navigate the defense acquisition process fast enough to matter is the entire bet.