The future of defence will not be decided only by who builds the best drone, the most advanced robot, or the most exquisite sensor. It will also be shaped by who can make all of those systems work together in a way that is operationally useful. That is the gap Arondite is trying to fill.
The London-based startup is building Cobalt, a software platform designed to connect robotic systems, autonomous assets, sensors, and operational data into a single interface. In plain terms, Arondite is building the connective tissue for an increasingly fragmented autonomous battlefield. Its premise is simple, but important: autonomy becomes far more valuable when it is orchestrated, interoperable, and kept under meaningful human control.
That positioning feels well timed. Defence is moving toward a world of mixed fleets, multi-domain operations, and software-defined capability. Militaries are buying more uncrewed systems, but those systems often arrive through different vendors, different standards, and different workflows. The result is familiar: silos, duplicated effort, and operators forced to piece together incomplete pictures under pressure. Arondite’s pitch is that the real bottleneck is no longer just the hardware. It is the layer that sits above it.
In a world full of autonomous systems, the strategic advantage may not come from owning the most platforms. It may come from making them work as one.
A company built around interoperability
Cobalt reflects that thesis clearly. Arondite describes the platform as the software layer between platforms, assets, robotic systems, autonomous systems, and data. It combines mission management, analytics and intelligence, a unified data model, and an integration suite. In practice, that means giving teams the ability to plan tasks that blend humans and machines, simulate options before execution, bring together sensor feeds and enterprise data, and deploy AI-enabled analytics in a controlled way.
What makes the company particularly interesting is its hardware-agnostic approach. Much of defence still gravitates toward closed ecosystems, where the software exists mainly to serve the hardware stack already owned by the prime or platform maker. That may simplify procurement in the short term, but it can also create lock-in and reduce flexibility. Arondite is taking the opposite route. Its value proposition is based on integrating third-party systems across land, sea, and air into one operational layer. In an environment where mixed fleets are the norm, that is not a cosmetic advantage; it is the product.
This is also why Arondite looks more like infrastructure than a point solution. If the defence stack is becoming more modular, then the software that lets different modules speak to each other starts to matter enormously. That is the space where Arondite wants to sit. Not as the company that owns every robot in the field, but as the company that makes a diverse robotic fleet usable at scale.
Why the team matters
Founder-market fit is one of the strongest parts of the Arondite story. CEO Will Blyth is a former British Army officer who later worked at the Ministry of Defence, Palantir, and Helsing. CTO Rob Underhill previously worked at BAE Systems and later became a founding engineer at Garrison Technology. That combination of operational credibility and defence-grade engineering is exactly what this category demands. It is hard to build mission-critical software for sensitive environments unless the founding team understands both the technical problem and the institutional reality of defence adoption.
That matters because selling into defence is rarely a straight line. Products do not win on elegance alone. They need to survive long procurement cycles, integration requirements, security standards, and the natural caution of end users. Arondite seems to understand this. Its early go-to-market appears built around pilots, partnerships, and integration into larger defence programmes rather than any fantasy of overnight disruption. In this market, credibility compounds slowly, but once embedded, it can become very durable.
Early signals of traction
There are already signs that the company’s thesis is resonating. Arondite raised $12.25 million across a pre-seed and seed round, with Index Ventures leading the seed financing and support from Concept Ventures, Creator Fund, and 9Yards Capital. For a young defence software company, that level of backing matters not only because of capital, but because it reflects confidence that the problem is both real and large.
More importantly, Arondite is beginning to appear in credible operational contexts. Babcock has said that Arondite’s Cobalt Operating System will serve as the autonomy and mission orchestration layer across ARMOR Force, its vision for a more hybrid Royal Navy capability. That does not make Arondite a category winner overnight, but it does suggest the company is moving beyond an abstract software story and into programmes where interoperability is a concrete requirement.
The real opportunity
Arondite is entering a competitive arena. Palantir, Anduril, and major defence primes all operate near parts of this stack, while governments may also try to build internally for sovereignty reasons. But Arondite’s opening is specific. Many incumbents still optimise around their own platform. Arondite is trying to optimise across platforms. If autonomy keeps spreading unevenly across air, land, sea, critical infrastructure, and public security, there will be growing demand for a trusted orchestration layer that is open, modular, and fast to integrate.
That is why the company feels aligned with broader structural shifts in defence. The market is moving away from isolated tools and toward integrated systems. The harder the operating environment becomes, the more valuable a shared operating picture becomes too. In that sense, Arondite is not just another autonomy startup. It is a bet on the importance of coordination itself.
The company is still early, and like many defence startups it will need to prove that pilots can convert into repeatable programmes and that software revenue can scale cleanly over time. But the core idea is compelling. In a world full of disconnected machines, the winner may not be the company that builds the most hardware. It may be the one that makes the whole system coherent.
Arondite is betting that the future of autonomy belongs to whoever can connect it. Right now, that looks like a smart bet.