Verda Secures €163M Series B to Expand Sovereign AI Neocloud Infrastructure

Helsinki-based startup Verda announced a €163 million (approximately $189 million) Series B funding round led by Emergence Capital, with participation from hardware vendor Supermicro, MUFG Innovation Partners, Varma Mutual Pension Insurance Company, and several infrastructure-focused angel investors. The company plans to use the capital to multiply its compute capacity over the next 12 months, expand platform services tailored to agentic AI workloads, and grow its footprint across Europe, the US, and Asia. This funding round underscores the surging demand for specialized "neocloud" infrastructure optimized entirely for machine learning and agentic workflows. For DevOps teams and ML platform architects, traditional general-purpose hyperscalers often impose unnecessary virtualization overhead, inflexible interconnect topologies, and high egress penalties. Verda's approach pairs direct bare-metal GPU clusters with native Kubernetes workload orchestration—such as Slurm integration via Slinky and Kueue—delivering high-efficiency distributed training and ultra-low-latency inference without the complexity of adapting traditional cloud environments. The development aligns with a broader shift in enterprise AI toward multi-cloud independence and sovereign AI infrastructure. As hyperscaler GPU allocations remain constrained and inference costs surge alongside the proliferation of autonomous agent systems, infrastructure startups that control the entire stack—from hardware procurement down to cluster scheduling—are capturing significant enterprise market share. Furthermore, with data sovereignty regulations tightening across the European Union, European enterprises are aggressively seeking regional compute providers that offer sustainable, low-carbon infrastructure alongside top-tier hardware. In practice, engineering leaders should evaluate neocloud alternatives like Verda when planning high-throughput model training pipelines or massive continuous inference deployments. While adopting a specialized AI cloud delivers price-performance advantages and reduced provisioning times, platform teams must account for operational trade-offs, including the need to maintain multi-cloud CI/CD pipelines, ensure custom IAM integration, and manage storage synchronization between primary enterprise data repositories and specialized GPU clusters.