Verda Raises €124M to Build European AI Cloud

There's a specific kind of startup move that tells you more than the press release does. It's not the headline raise — it's what comes right after. Verda, the Helsinki-based AI infrastructure company formerly known as DataCrunch, raised €100 million ($117 million) in April 2026. Then, just months later, it secured an additional €22 million loan from the Nordic Investment Bank (NIB). That's roughly €124 million in fresh capital in less than a year, and the sequencing matters.

The April round — reported by EU-Startups, tech.eu, and ArcticStartup, among others — was equity-based, backed by Lifeline Ventures, Tesi, Varma, and byFounders, according to technews180.com. The July NIB loan is debt. That's a different instrument with a different logic, and founders who read these announcements as interchangeable funding news are missing something important about how Verda is actually capitalizing itself.

Why the NIB Loan Is the Interesting Part

Development bank loans aren't typical startup financing. The Nordic Investment Bank is a multilateral institution owned by the Nordic and Baltic states. When it lends to a private company, it's doing so because the investment aligns with some version of regional economic or environmental policy — in this case, presumably the development of European AI capacity and the deployment of energy-efficient compute infrastructure. The NIB doesn't take equity, doesn't sit on your board, and generally charges favorable rates compared to commercial debt. It also signals something to the market: that a credible institutional actor has looked at Verda's business and decided it's worth backing with public-adjacent capital.

That's a form of validation that's structurally different from venture dollars. VCs make concentrated bets on outliers and are explicitly in the business of accepting high failure rates. Development banks are underwriting more like an infrastructure lender — they want to know the thing will actually get built and used. The combination of private equity and development bank debt in Verda's capital stack suggests the company is pitching itself — credibly, to multiple audiences — as both a high-growth startup and a piece of regional infrastructure.

Whether that dual framing is sustainable is the open question.

The Nordic Power Thesis

Verda's proposition, as covered by tech.eu, is built explicitly on clean Nordic power. That's not just marketing. AI compute is obscenely energy-hungry — training large models and running inference at scale requires data centers that consume electricity on a utility scale. Where that power comes from increasingly matters, both for sustainability optics and for operating cost. The Nordic region's combination of hydropower, wind, and cold ambient temperatures (which reduce cooling costs) gives infrastructure companies operating there a genuine structural advantage, not a made-up one.

This is the part of Verda's pitch I find most defensible. The hyperscalers — AWS, Azure, Google Cloud — have been racing to build out AI compute capacity globally, but they're constrained by power availability everywhere. A European provider with reliable, low-carbon energy sources and proximity to European enterprise customers has a real wedge. The EU's AI Act and general regulatory pressure around data sovereignty adds another layer: European companies that want to run sensitive workloads on European soil, under European law, are an underserved market that Verda is directly positioned to capture.

Where the Expansion Money Goes

According to EU-Startups, Verda planned to hire more than 100 people by the end of 2026 alongside its April raise. theaiworld.org notes the expansion targets the UK, US, and Asia. technews180.com confirms the geographic scope: European AI cloud platform expanding into the US and UK.

That's an ambitious geographic spread to pursue simultaneously. The UK and US are not the same market. The US market is dominated by hyperscalers with enormous scale advantages and deeply entrenched enterprise relationships. Competing there on cost is nearly impossible; competing on sovereignty, latency, or specific regulatory compliance use-cases is at least theoretically viable. The UK is a different shape — closer to European data governance norms post-Brexit than the US, with a specific AI policy environment that's been friendlier to infrastructure investment than the EU's more prescriptive approach.

Asia is a separate thesis entirely, and the sources don't specify which Asian markets Verda is targeting. That matters enormously — Japan and Singapore are very different bets from India or South Korea. The record is thin here, and I won't invent detail the sources don't support.

The Rebrand Signal

The fact that Verda was formerly DataCrunch is worth pausing on for a moment. Rebrands cost money, and they carry risk — you're voluntarily erasing whatever brand equity you'd accumulated. Companies typically rebrand for one of two reasons: they're trying to escape a reputation problem, or they're signaling a significant strategic pivot to a new market or identity. Nothing in the available record suggests a reputation problem. The more plausible read is that "DataCrunch" felt like a technical, developer-adjacent name, while "Verda" reads as enterprise-ready — cleaner, more neutral, easier to internationalize. That's a deliberate signal about who the company is now selling to, and it tracks with the expansion into UK and US enterprise markets.

Renaming yourself right before a major raise and international expansion is a coherent playbook. It's also the kind of thing that gets dismissed as cosmetic when it's actually load-bearing strategic communication.

The Capital Structure Question Founders Should Ask

Here's the thing about stacking equity and development bank debt in rapid succession: the equity round's liquidation preferences are now senior to the new company debt in some scenarios, but the dynamics get complicated depending on the loan terms. NIB debt gets repaid before equity distributions in a wind-down — that's standard — but the equity investors' liquidation preferences mean that in an exit, the VCs get paid before common shareholders regardless. For the 100-plus employees Verda is hiring, this is the unsexy math that determines whether their equity is worth anything in a downside scenario.

None of this is Verda-specific. It's the standard architecture of venture-backed startups with institutional debt. But as someone who writes about this for founders and employees who deserve to know what they're signing: if you're taking a job at Verda on the promise of meaningful equity participation, you should ask what the liquidation preference stack looks like across all these instruments. The total raise sounds impressive. What matters is the waterfall.

What Makes This Worth Watching

Verda is doing something specific and testable: betting that European AI infrastructure demand is large enough, and that European buyers are differentiated enough from their US counterparts, to support a scaled independent provider rather than just deferring everything to AWS or Azure. The NIB backing adds credibility to the thesis that this isn't purely a venture return play — there's a genuine infrastructure-development argument underneath the startup framing.

The next few quarters will clarify whether the US and UK expansion is real customer traction or a slide in the deck. Hiring 100-plus people while simultaneously entering multiple new geographies on a single capital raise is the kind of plan that works beautifully when growth outpaces burn and looks catastrophic when it doesn't. The debt component adds a repayment obligation that equity doesn't — which focuses minds in useful ways.

Europe has been trying to build a credible alternative AI infrastructure layer for years. Most attempts have either been too small to matter or too slow to compete. Verda has more capital and a cleaner strategic thesis than most. Whether that's enough is the question the next eighteen months will answer.

---

By Alex Volkov, Startup Ecosystem & Venture Capital Reporter, Buzzrag