Mistral Backs Italian Startup Clastix with $3.1M to Challenge Cloud Giants

·
Listen to this article~4 min

Clastix, a Naples-based Kubernetes infrastructure startup, raised $3.1M in Seed funding led by CDP Venture Capital with Mistral backing. The company aims to build sovereign infrastructure for Europe's AI ecosystem.

### From Naples to the Front Lines of Europe's AI Infrastructure Clastix, a Naples-based infrastructure software company, just raised €2.9 million ($3.1 million) in a Seed round. It's their first outside funding, and it's led by CDP Venture Capital Sgr. Mistral and Vertis SGR also joined in through the Vertis Venture 6 Digital Sud fund, which is partly backed by the EU's Next-Generation EU program. This isn't just another funding story. It's a bet on Europe's digital sovereignty—and on the idea that the next wave of AI won't be won by models alone, but by whoever controls the infrastructure underneath them. ### The Problem: Kubernetes Can Be a Money Pit Kubernetes is the go-to system for running containerized apps at scale. But managing it across hundreds or thousands of machines? That's where things get expensive and complicated. Clastix was founded in 2020 by Adriano Pezzuto and Dario Tranchitella, two engineers who'd spent years wrestling with Kubernetes inside big companies. They knew the pain firsthand. Their first open-source project, Capsule, brought multi-tenancy to Kubernetes—letting multiple teams share the same cluster safely. Then came Kamaji, which runs tenant control planes as pods on shared infrastructure instead of dedicated machines. Clastix calls this eliminating the "control plane tax." Together, these became kMetal, their commercial platform for fleet-scale Kubernetes operations. It's already in production with AI labs, telecoms, cloud providers, and the public sector. Customers include Fastweb, NVIDIA, and Mistral. > "The next phase of AI won't be decided by models alone but by who controls the infrastructure layer beneath them." – Adriano Pezzuto, co-founder and CEO of Clastix ### Why Mistral's Backing Matters Mistral, the French AI powerhouse, isn't just an investor—they're a customer. Timothée Lacroix, Mistral's co-founder and CTO, said Clastix is "the backbone of our Kubernetes infrastructure." He added that their work on multi-tenancy and managed control planes helps Mistral run training and inference clusters at scale. That's a strong endorsement. It also signals a shift: European AI companies want alternatives to hyperscaler-dependent infrastructure. They want to keep control of their hardware and data while still getting hyperscaler-grade operations. ### What's Next for Clastix The new capital will go toward product development, customer success, solutions engineering, and go-to-market expansion. Over the next three years, Clastix plans to grow its team to meet rising demand and serve larger enterprise customers. They'll also keep investing in Kamaji, their open-source Hosted Control Plane engine. It will remain free under the Apache 2.0 license—unchanged. - **Product development:** Expand R&D to support AI-scale workloads. - **Customer success:** Help enterprises get the most out of kMetal. - **Go-to-market:** Bring Clastix into new markets as they build AI infrastructure. ### The Bigger Picture Europe has been trying to build its own tech sovereignty for years. Clastix is a concrete example of that ambition. By offering an alternative to hyperscaler-dependent infrastructure, they're giving European enterprises and AI organizations a way to run Kubernetes at scale without handing over control to a handful of cloud giants. As Alessandro Scortecci, Chief Investment Officer at CDP Venture Capital, put it: "This investment reflects CDP Venture Capital's commitment to backing strategic technologies developed in Italy that can strengthen Europe's technological sovereignty and long-term competitiveness." It's a small round, but the implications are big. If Clastix succeeds, it could become the infrastructure layer that Europe's AI ecosystem runs on—and that's something worth watching.