Omen AI Raises $31 Million to Bring Real-Time Fluid Monitoring to Data Centers

Data centers are under intense pressure to wring more performance out of every rack of GPUs, but one of the industry’s quietest failure points is getting less attention than it deserves: the fluids circulating through cooling systems. As operators push liquid-cooled hardware harder, they also raise the risk of bacterial growth, contamination, and costly shutdowns. Omen AI thinks it has built a way to spot those problems before they spiral into downtime.

The startup has announced a $31 million Series A round as it pivots from industrial equipment monitoring into a fast-growing corner of AI infrastructure. Instead of waiting for fluid samples to be shipped to a lab, Omen is selling a compact spectrometer that analyzes coolant and other fluids continuously on site, giving operators a live read on chemistry that can affect temperature, equipment health, and uptime.

The company’s wager is straightforward: as AI workloads become more power-hungry, the systems that keep compute running cool and reliable will matter as much as the chips themselves. If a contamination event can take a rack offline for hours, the economics of prevention become compelling very quickly.

Why data center cooling is suddenly a bigger business problem

AI has changed the data center from a facility that mostly housed servers into one that behaves more like a tightly tuned industrial plant. Higher-density GPU deployments generate more heat, and the industry has increasingly turned to liquid cooling to keep temperatures under control. Liquid systems are more efficient than air at moving heat away from dense hardware, but they introduce a different kind of operational risk: the fluid itself can degrade.

In many systems, the coolant is a blend of water and additives designed to discourage bacterial growth and protect the machinery. The balance matters. More water improves heat transfer, but it can also make the environment easier for microbes to thrive in. Once bacteria start to accumulate, they can clog the flow, reduce cooling performance, and create the kind of contamination headache that operators would rather avoid entirely.

The traditional response is reactive. Teams take samples, send them to a lab, and wait for results. If a problem is discovered late, the fix may require flushing the loop, a process that can force a rack offline for five or six hours. In a hyperscale or AI training environment, that kind of interruption can be extremely expensive.

Omen’s pitch is that data center operators should not be guessing about coolant health or relying on delayed lab work to make decisions. They need an always-on signal from the fluid itself.

From heavy equipment to AI infrastructure

Omen did not start by selling into data centers. The company was initially built around industrial machinery, where fluid monitoring can help identify wear before it becomes a breakdown. Early customers included Caterpillar dealerships, which gave the startup a foothold in heavy equipment markets and access to operators that already understood the value of predictive maintenance.

That original direction turned out to be only the beginning. As those dealership relationships expanded, they started pointing Omen toward another environment full of pumps, liquids, and critical uptime requirements: buildings, especially those supporting high-value equipment and power systems.

That shift became more pronounced as data center operators and their suppliers began to ask whether the same sensing approach could be applied to on-premises power equipment and cooling infrastructure. In other words, Omen discovered that a business designed to watch fluids in construction machinery could also help watch the fluids inside AI facilities.

The broader logic of the product stayed the same. Whether the system is a bulldozer, turbine, or data center cooling loop, fluid analysis can reveal the early signs of failure. The difference is the scale and urgency of the market now coming into focus.

How Omen’s sensor approach works

At the heart of Omen’s product is a small spectrometer that analyzes the chemical state of fluid in real time. Spectroscopy, broadly speaking, uses light to identify what is present in a sample. In industrial settings, that can be useful for detecting not just contamination but also trace materials that suggest hardware wear.

Laberge says the device can identify bacterial growth before the problem becomes severe enough to disrupt operations. It can also look for signs that components are wearing down. If the system detects copper or chromium, that may point to pump wear. If it detects silicon, that may indicate seals are deteriorating.

The point is not merely diagnosis after the fact. It is to move fluid monitoring from a periodic checkup model to a continuous, predictive one. That can help operators schedule maintenance before failures cascade into downtime, and it can also help them operate closer to the edge without crossing it blindly.

Two technical trends have helped make this approach more practical: cheaper optical hardware and better signal-processing software. Together, they let Omen build a sensor that is small enough and affordable enough to deploy at scale while still producing meaningful readings from noisy industrial environments.

“You’re not risking huge amounts of downtime because you have no insight into what’s going on chemically,” Omen CEO and founder Zach Laberge said, emphasizing the operational value of real-time visibility.

Laberge also argued that the economics have finally reached a point where the idea makes sense outside of a lab. Cheaper hardware lowers the barrier to deployment, while signal processing improves the quality of the data operators receive.

Inside the funding round

Omen said it has raised $31 million in a Series A round led by Nava Ventures. The financing also included participation from CRV, Vanderbilt University, Mann+Hummel, Starhill Holdings, and Hard Launch Capital. In addition, executives from companies including Bridgestone, General Motors, Johnson Controls, and Tensorwave made personal investments.

That mix of investors is notable. It suggests Omen is attracting not just software-oriented venture capital but also strategic interest from industrial and infrastructure players that understand the value of monitoring critical systems. For a company positioned at the intersection of hardware, industrial maintenance, and AI infrastructure, that is a meaningful signal.

The latest round brings Omen’s total funding to $40 million since the company was founded in 2024. That pace of capital formation reflects both investor appetite for AI infrastructure tools and the broader search for businesses that can profit from the knock-on effects of the AI buildout.

Data centers are not the only piece of the AI economy attracting attention, but they are one of the clearest examples of a supply chain that is being reshaped by demand for compute. Anything that helps operators increase uptime, reduce maintenance risk, or improve efficiency is likely to find a receptive audience.

A young founder with unusual traction

Laberge’s background is a central part of Omen’s story. He founded his first startup in 2020 at the age of 14, eventually raising $3 million for a business that installed sensors on construction equipment. He later left high school to pursue entrepreneurship full time, with support from his parents, including his mother, who previously served as Ontario’s minister of education.

That early start matters because it helps explain both the company’s technical orientation and the trust he appears to have built with larger industrial customers. Omen’s supporters argue that Laberge is unusually credible for a founder so early in his career.

Cory Rellas, a partner at Nava Ventures and a member of Omen’s board, said the company benefited from introductions to major customers that quickly validated its approach, adding that it is uncommon to see such a young founder earn respect from large, established corporations in a traditionally slow-moving sector.

That validation has become especially important as Omen broadens from construction and heavy equipment into data centers. Industrial buyers tend to be conservative, and infrastructure customers often want proof that a tool works before they commit. In that sense, the company’s early work with equipment and dealerships may have given it a credibility bridge into a much larger market.

Why data center operators care about fluid health

To understand why Omen is getting attention, it helps to look at the operational stakes inside a modern AI facility. A large compute installation may contain hundreds or thousands of GPUs, each drawing substantial power and producing heat that must be removed continuously. Cooling is not an auxiliary feature; it is part of the core operating system of the building.

That makes every part of the thermal loop important. Pumps, seals, coolant chemistry, and filtration all play a role in whether a facility can stay online at full capacity. If one part of the system degrades, it can affect the rest.

In conventional setups, fluid testing is often periodic and manual. An operator may collect a sample, send it out, and wait for results. That may be adequate when systems are relatively stable or when the cost of missing a problem is modest. But in AI infrastructure, where uptime translates directly into revenue and research throughput, slow feedback can be too little, too late.

Omen’s real-time monitoring promise is attractive because it gives operators a chance to intervene before a rack has to be taken offline. Even a few hours of avoidance can matter when the value of the underlying compute is measured in millions of dollars.

The costs of a flush-and-restart approach

When a coolant loop becomes contaminated, flushing the system may be the only safe option. But flushing is a blunt instrument. It disrupts operations, consumes labor, and can force expensive hardware to sit idle while the system is cleaned and refilled.

For a rack supporting AI workloads, the cost of a five- or six-hour outage is not only the direct loss of compute time. There is also the scheduling friction, the potential missed deadlines for training jobs, and the operational complexity of restarting work across a distributed cluster.

That is why a monitoring layer that spots early warning signs can have value well beyond the price of the sensor itself. It changes maintenance from emergency response to planning.

Who is buying into the vision

Omen says it is already working with a dozen data center customers as it builds out its product. Among them is TensorWave, a company building an AI compute cloud on AMD chips.

TensorWave president Piotr Tomasik said the coolant running through large systems is a critical variable that much of the industry still does not monitor closely enough, and described better infrastructure monitoring as central to supporting AI customers.

That endorsement points to the kind of market Omen is targeting. TensorWave is not just a generic infrastructure buyer; it is part of the broader AI compute ecosystem that is racing to bring more capacity online. If companies like that see utility in live fluid analytics, the market opportunity extends beyond a narrow industrial niche.

Other potential customers include operators and suppliers around power generation, building systems, and thermal management. The overlap between industrial maintenance and data center operations may become one of Omen’s most valuable advantages. The same sensor platform can potentially serve multiple environments with fluid-dependent equipment.

Competition is emerging, but the market is still early

Omen is not the only company thinking about fluid analytics for data centers. Earlier this month, Pyxis, a more established water-monitoring company, introduced a coolant monitoring product aimed at data centers. That suggests the category is beginning to attract attention from firms with existing measurement and water-quality expertise.

Still, the market appears early enough for multiple approaches to coexist. Some companies may focus on broader water systems, while others, like Omen, are building specifically around the chemistry and operational needs of AI infrastructure. As liquid cooling becomes more common, there is room for specialized tools that provide either general monitoring or more granular diagnostic capability.

There is also a larger strategic question. As data centers evolve, their owners may increasingly want sensors that not only detect problems but also integrate with automation, maintenance workflows, and facility management software. The winner in that space may be the company that can turn chemical data into actionable operational intelligence, not just raw readings.

The role of liquid cooling in the AI buildout

The rise of liquid-cooled chips is one of the most consequential engineering shifts in the AI infrastructure boom. As compute density rises, air cooling alone becomes less effective in many environments. Liquid systems can remove heat more efficiently, enabling tighter packing and higher performance.

But each engineering improvement introduces its own maintenance burden. More advanced thermal systems demand more sophisticated monitoring. That opens the door for companies that can observe not just temperatures and flow rates but the health of the fluid itself.

In that sense, Omen sits in a broader wave of infrastructure tooling that is being created because AI workloads are pushing old assumptions to their limits. The opportunity is not limited to chips, servers, and networking. It extends to pumps, pipes, seals, and coolants — the less glamorous but equally essential parts of the stack.

Why the chemistry matters

Coolant is not just water moving through a pipe. It is a carefully engineered medium whose composition affects thermal performance, biological stability, and equipment longevity. If operators change the mix to improve heat absorption, they may be trading one kind of efficiency for another kind of risk.

That tradeoff is exactly where sensors become useful. Instead of forcing operators to choose between performance and safety without enough information, continuous analysis allows them to tune systems more intelligently.

The promise is not that contamination will disappear. It is that problems will be found early enough to manage them without major disruption.

What the funding signals about the market

Omen’s new capital raise arrives at a time when investors are searching for picks and shovels businesses tied to AI infrastructure. The hyperscalers may capture most of the headlines, but the supporting layers of the stack are where many durable businesses are likely to emerge.

Fluid monitoring may sound niche, but the category sits at the intersection of a few powerful trends: expanding compute demand, growing liquid-cooling adoption, pressure to reduce downtime, and the shift toward predictive maintenance.

For investors, that combination is attractive because it links a relatively narrow product to a very large and expanding customer base. Every new AI data center, every higher-density rack, and every on-premises power system potentially expands the addressable market.

Strategic investors from industrial and infrastructure sectors also appear to see the benefit. Their participation suggests the technology could have relevance beyond a single use case, extending across manufacturing, energy, building systems, and heavy equipment.

Key facts at a glance

Timeline: how Omen moved toward data centers

What happens next

Omen now faces the challenge familiar to most infrastructure startups: turning a strong technical idea into a repeatable deployment model. Data center customers can be demanding, and they often want evidence that a tool will integrate cleanly with existing systems and pay for itself quickly.

But the company appears to have several advantages. It has early traction with a dozen customers, support from recognized investors and strategics, and a product that addresses a very concrete pain point in an industry under pressure to expand rapidly. It also benefits from a founder who has already shown an ability to attract serious industrial customers despite his age.

Whether Omen becomes a major player will depend on execution, but the timing is favorable. AI infrastructure is growing fast, liquid cooling is becoming more important, and the operational cost of downtime is high enough to justify new monitoring tools. In that environment, a spectrometer that can warn operators before a coolant loop turns into a crisis may be exactly the kind of unglamorous but essential product the market rewards.

For now, Omen is betting that the future of AI infrastructure will be won not only by better chips and larger clusters, but also by the ability to understand what is happening inside the fluids that keep those systems alive.

superintelligencenews@gmail.com