Clastix raises €2.9M to expand its commercial Kubernetes platform

Why it matters

Clastix's seed finances a shift from an open-source Kubernetes control-plane tool to a broader commercial platform. Mistral's role as both investor and user gives the AI-infrastructure pitch a concrete customer connection, while kMetal's success will depend on whether operators pay for the larger product around Kamaji.

Adriano Pezzuto, Clastix's co-founder and CEO, has spent years working on the operational problem behind the company's pitch: large organizations can end up running many Kubernetes clusters, each with its own control plane to operate. On September 24th, Tech.eu reported that Clastix had raised a 2.9 million euro seed round to expand its team and develop its Kubernetes platform.

The round was led by CDP Venture Capital, with participation from Mistral and Vertis SGR through its Vertis Venture 6 Digital Sud fund. Mistral is also a user: its co-founder and CTO, Timothee Lacroix, told Tech.eu that Clastix supports Mistral's Kubernetes infrastructure, including multi-tenancy and managed control planes for training and inference clusters. For Clastix, that makes Mistral both an investor and a customer validating the infrastructure it wants to sell.

Clastix describes its origins as an engineering response to work its founders had done inside large enterprises. They had managed Kubernetes platforms at scale and encountered the operational burden firsthand, the company says. Their early work produced Capsule, an open-source multi-tenancy project, and later Kamaji, which runs tenant Kubernetes control planes as pods within a shared management cluster instead of on dedicated machines.

That product history now leads to the commercial bet. Kamaji will remain open source under the Apache 2.0 license; Clastix is building kMetal as the paid management layer around it. Clastix's own account describes kMetal as a broader platform for bare-metal Kubernetes, combining hosted control planes with fleet management and enterprise operating capabilities. The distinction gives infrastructure teams a path from adopting an open-source component to buying a more complete platform without Clastix abandoning the project that introduced it.

From control planes to the whole platform

Kamaji addresses a specific source of cost as cluster counts grow. Kubernetes control planes run the components that manage a cluster. Traditionally, each cluster needs dedicated machines for those components, adding hardware and operational work as teams create more clusters. Kamaji runs tenant control planes as workloads on shared infrastructure instead. Clastix's technical documentation describes the model; the claimed efficiency benefit depends on how an operator configures and uses its infrastructure.

Clastix has been extending that idea beyond control-plane hosting. In March, Pezzuto outlined the move from Kamaji to kMetal, arguing that hosted control planes address only part of the cost of running multi-tenant Kubernetes. The kMetal plan brings together infrastructure management and isolation as well as the control plane. On August 31st, Clastix announced kMetal 1.0's general availability, positioning it for enterprises running Kubernetes directly on bare metal.

That timing gives the seed a concrete job: fund the commercial product's expansion after its general-availability release, while maintaining the open-source engine underneath. Tech.eu reported that Clastix plans to spend on product development, customer success, solutions engineering and go-to-market work for kMetal, as well as research and development for AI workloads. The company also expects to grow headcount over the next three years, but did not attach a hiring target to that plan.

Pezzuto framed the AI work as an infrastructure bet. "The next phase of AI won't be decided by models alone but by who controls the infrastructure layer beneath them," he told Tech.eu. The claim fits Clastix's existing product direction: its focus is the systems used to provision and operate clusters, not AI models or applications. Whether AI infrastructure becomes a meaningful new market for Clastix will depend on customers choosing its platform for the underlying operational work.

A customer-investor with a direct stake

Mistral's participation sharpens the logic of the round. Lacroix said Mistral uses Clastix's work to operate training and inference clusters at scale, while describing the investment as support for Europe's AI infrastructure. That is a specific endorsement from a prospective source of market credibility, though it does not establish how widely the products are deployed or what revenue Clastix earns from them.

Tech.eu said Clastix's technology is in production with infrastructure teams in Europe and North America, across AI labs, telecommunications, cloud services and the public sector. The report gave no customer count or revenue figure. The round's value is therefore clearer as a financing and product milestone than as evidence of commercial scale: CDP Venture Capital led a 2.9 million euro seed, with Mistral and Vertis participating, as Clastix shifts from an open-source control-plane engine toward a broader commercial platform.

The strategic test is whether kMetal can persuade enterprise operators to buy the larger operating layer while Kamaji continues to attract users and developers on its own. Clastix has kept the two products linked: Kamaji supplies the hosted-control-plane capability, while kMetal packages that foundation with fleet operations. The seed backs Pezzuto's effort to make that transition work, and its AI positioning puts the company in a market where infrastructure capacity and cluster operations are becoming part of the product, not just background plumbing.