Inside Nebius: The Dutch Rooted AI Cloud Builder
A New Kind of AI Infrastructure Company
Nebius emerged quietly, almost unexpectedly, as one of Europe’s most strategic AI-infrastructure players. Its corporate home in Amsterdam–Schiphol Rijk gives it a European identity, but its operations stretch far beyond Dutch borders. Born from the international restructuring of Yandex in 2024, Nebius now positions itself as an independent global technology group, building cloud systems designed not for traditional enterprise workloads, but for intensive AI training and high-performance computing.
Where classic cloud providers evolved from managing servers, Nebius enters the market from the opposite direction: it builds infrastructure around GPUs first and cloud services second. That mindset shapes how they design datacenters, choose locations and engage with partners. Their facilities in Finland, Iceland, the United States and Central Europe are not generic compute warehouses; they are purpose-built for AI acceleration, high-density racks and thermal management challenges that only large-scale model training brings.
Why These Locations Matter
Nebius has become particularly strategic about geography. Helsinki, for example, offers naturally cold air, allowing efficient cooling. Iceland provides not only geothermal and renewable electricity but also an environment where energy-hungry AI clusters can operate with a markedly lower carbon footprint. In the United States—New Jersey and Kansas City—the company responds to growing demand from American enterprise customers who want AI compute within domestic borders.
Each region demonstrates a different philosophy. Finland represents environmental optimisation, Iceland is a bet on clean energy abundance and the US clusters reflect Nebius’s ambition to compete directly with hyperscalers in one of the most competitive markets on earth. The company is expanding quickly, but it is doing so by selecting places where climate, energy infrastructure and industry partners converge in the right way.
The Energy Question: A Reality and an Opportunity
Training AI models requires enormous power and cooling capacity—an unavoidable reality of modern compute. Nebius does not hide from this; instead, it frames efficiency as an engineering challenge. Their newer facilities are designed around high-density GPU clusters that operate closer to the limits of what traditional data halls can support. The company leans heavily on renewable-energy regions, advanced heat-reuse concepts and innovations in liquid cooling.
The environmental impact of AI infrastructure is significant, but Nebius attempts to show that responsible design can coexist with rapid expansion. In Finland and Iceland, much of the electricity supporting Nebius datacenters comes from renewable sources. And in continental Europe, they increasingly explore ways to integrate facilities into local energy ecosystems so that waste heat becomes an asset and not a by-product.
The story is not one of perfection, but of constant optimisation—a message that resonates with regulators, investors and the communities around their sites.
People Behind the Machines
Nebius is not only scaling machines; it is scaling talent. As the scope of operations grows, the company searches for people who understand the intersection of cloud architecture, AI engineering and large-scale infrastructure. Their hiring focus tends to cluster around high-performance compute specialists, GPU cluster engineers, distributed systems developers and networking experts who can design low-latency fabrics for AI workloads.
Beyond the technical core, Nebius increasingly attracts sustainability engineers, energy-systems analysts, datacenter operations teams and specialists in cooling, resilience and power distribution. Because the company operates in multiple jurisdictions, it also needs legal, compliance and security professionals who understand the regulatory differences between the EU, the US and emerging markets.
In many ways, Nebius offers the identity of a startup with the complexity of a hyperscaler: a rare combination that appeals to engineers who want to solve difficult infrastructure problems without getting lost in the bureaucracy of larger cloud providers.
Why Nebius Matters Now
The global AI sector is accelerating and the bottleneck is no longer software—it is compute. Nebius understands that the next wave of AI progress depends on access to scalable GPU infrastructure, clean energy and datacenters that can support the extreme densities required for training next-generation models. By anchoring its corporate home in the Netherlands and building sites across renewable-energy corridors, Nebius shows how a European-rooted company can influence the future of AI infrastructure on a global scale.
Nebius is still young compared to giants like Google Cloud, AWS or Microsoft Azure, but it is moving fast. It has secured major partnerships, expanded its footprint and demonstrated that new players can redefine how and where AI datacenters are built. In a market dominated by a few hyperscalers, Nebius stands out by choosing a different path—one that blends engineering ambition with environmental pragmatism and a clear understanding of the geopolitical landscape of compute.
