Armada’s $230M Series B: Building the U.S. Sovereign Edge AI Stack

The Announcement

Armada, a modular edge AI infrastructure company, has closed a $230 million Series B at a $2 billion pre-money valuation, in a round described as heavily oversubscribed and co-led by Overmatch, BlackRock, and 8090 Industries. Simultaneously, the company announced a Global Framework Agreement with Johnson Controls to produce modular data centers at a new 400,000-square-foot facility in Arizona, branded Galleon Forge One, with production of its megawatt-scale Leviathan units expected to begin this summer. The raise brings Armada’s total funding to nearly $500 million. Customer bookings grew 540% from FY25 to FY26, with Q1 FY27 reporting a 2,000% increase year over year, driven by deployments spanning U.S. Navy maritime exercises, offshore oil rigs in Norway, and renewable energy compute in Australia.

The Bigger Picture

The Edge AI Infrastructure Gap Is Real and Widening

The conventional AI infrastructure conversation has been dominated by hyperscale data centers: massive, fixed facilities that require years of permitting, construction, and grid negotiation before a single GPU processes a single token. That model is increasingly inadequate for a growing category of workload, namely AI inference and training that must happen where data is generated, not where it is convenient to build.

Armada is betting that the next phase of enterprise AI adoption will be defined not by who has the most GPUs in a centralized facility, but by who can deploy capable, sovereign compute to the most constrained and demanding environments. The use cases in this announcement make that argument concretely: a defense partner went from infrastructure requirement to operational capability in six days; the U.S. Navy ran mission applications on a vessel during a multinational exercise; and Aker BP is moving toward autonomous operations on an offshore rig in the Norwegian Continental Shelf. None of those deployments were possible through a hyperscaler console.

ECI Research’s 2025 AI Builder Summit survey found that 59% of organizations are investing in Agentic AI for IT Operations today. Much of that investment assumes reliable, low-latency compute in the environment where operations actually occur. For sectors like defense, energy, maritime, and remote industrial, “the environment” is not a cloud region.

What This Means for ITDMs

The economics here deserve close attention. Armada’s modular approach could address a structural problem that most enterprise infrastructure planning ignores: the cost and time of building compute capacity at remote or sovereign sites is not linear with hyperscale economics. A traditional data center buildout in a regulated, remote, or geopolitically sensitive environment can take three to five years and require grid capacity that simply does not exist. Armada’s Leviathan units are designed to be manufactured at scale and deployed on demand, which compresses that timeline dramatically.

For ITDMs in defense, energy, utilities, and critical infrastructure, the calculus is shifting. The question is no longer “can we run AI in the cloud?” but “what portions of our AI workload must stay local, must run without connectivity, or must remain under domestic jurisdiction?” That is a governance and risk question as much as a technical one. The Johnson Controls partnership strengthens the thermal management and facilities expertise that makes large-scale modular deployment reliable in production, not just in a proof of concept.

The $2 billion pre-money valuation and the investor composition (BlackRock, NightDragon, Mitsui, Singtel Innov8 alongside defense-aligned funds) signals that institutional capital views sovereign edge AI infrastructure as a durable category, not a niche. ITDMs evaluating multi-year AI infrastructure strategies should treat Armada as a credible tier-one option for sovereign and edge workloads, alongside rather than instead of their hyperscaler relationships.

What This Means for Developers

From a technical standpoint, the Armada stack is worth examining for several reasons. The Leviathan units are designed for high-density AI training and inference, which means they are targeting the GPU-intensive workloads that currently bottleneck most enterprise AI pipelines. The company’s Marketplace, which integrates partnerships with Microsoft, NVIDIA, Palantir, and Dell Technologies, suggests a platform play: Armada wants to be the substrate on which familiar AI development toolchains run, not a proprietary dead end.

For developers building AI applications that touch regulated data (health records, defense logistics, financial transactions), the sovereignty argument matters at the architecture level. Running inference on a Leviathan deployed on-premises or in a controlled sovereign environment changes the threat model substantially compared to sending data to a shared cloud inference endpoint. ECI Research’s 2025 AI Builder Summit survey found that 44% of enterprise AI leaders have only moderate confidence that AI agents can act autonomously without human intervention. That confidence gap is partly a reliability question, but it is also a data governance and auditability question. Edge-local compute with controlled egress is one architectural response to that concern.

The Aker BP deployment is a useful mental model for developers thinking about agentic workloads at the edge: AI models running on the rig, processing sensor data in real time, with the goal of enabling autonomous drilling operations. The orchestration, retraining, and monitoring challenges for that use case are materially different from a cloud-native SaaS application.

What’s Next

Manufacturing Velocity Will Define the Category

The Galleon Forge One facility in Arizona is the pivotal near-term variable. Continuous production beginning this summer means Armada will face its first real test of whether the manufacturing model can match the sales velocity implied by 2,000% bookings growth in a single quarter. A 400,000-square-foot facility producing megawatt-scale modular data centers is an industrial undertaking that introduces supply chain, quality, and logistics complexity that is categorically different from software scaling.

ECI Research’s 2025 report on cloud and enterprise strategy noted that organizations with the highest FinOps maturity are distinguished not by the most advanced tools, but by the most integrated teams. The same principle applies to hardware manufacturing at scale. Armada’s ability to integrate Johnson Controls’ thermal and facilities expertise with its own edge computing software stack, across a high-volume production environment, will determine whether its deployment promises hold up in practice. Investors and enterprise customers should watch Q1 FY28 bookings fulfillment rates as a leading indicator of execution quality.

The Sovereign AI Infrastructure Market Is Consolidating Early

The investor composition of this round (a sovereign wealth-adjacent asset manager in BlackRock, a defense-focused fund in NightDragon, and strategic partners in Mitsui and Singtel Innov8) reads as a deliberate effort to build a globally deployable sovereign AI infrastructure network before the market fully consolidates. Armada is not the only company pursuing modular edge compute, but it is one of the few that combines AI-native software, high-density GPU infrastructure, defense-grade deployment experience, and a manufacturing partnership at scale.

We expect Armada’s next 18 months to focus on three priorities: proving the Galleon Forge One production model at volume, deepening the partner Marketplace to increase application density on deployed units, and expanding the geographic footprint through Mitsui and Singtel Innov8’s networks in Asia-Pacific. Enterprise buyers in regulated industries who have been waiting for the modular edge AI market to mature should treat this announcement as a maturity signal. The infrastructure category is moving from bespoke deployments to production at scale, and the procurement window for favorable early-adopter positioning is shortening.

Authors

  • With over 15 years of hands-on experience in operations roles across legal, financial, and technology sectors, Sam Weston brings deep expertise in the systems that power modern enterprises such as ERP, CRM, HCM, CX, and beyond. Her career has spanned the full spectrum of enterprise applications, from optimizing business processes and managing platforms to leading digital transformation initiatives.

    Sam has transitioned her expertise into the analyst arena, focusing on enterprise applications and the evolving role they play in business productivity and transformation. She provides independent insights that bridge technology capabilities with business outcomes, helping organizations and vendors alike navigate a changing enterprise software landscape.

    View all posts
  • Paul Nashawaty

    Paul Nashawaty, Practice Leader and Lead Principal Analyst, specializes in application modernization across build, release and operations. With a wealth of expertise in digital transformation initiatives spanning front-end and back-end systems, he also possesses comprehensive knowledge of the underlying infrastructure ecosystem crucial for supporting modernization endeavors. With over 25 years of experience, Paul has a proven track record in implementing effective go-to-market strategies, including the identification of new market channels, the growth and cultivation of partner ecosystems, and the successful execution of strategic plans resulting in positive business outcomes for his clients.

    View all posts