Summary
- Omen AI has raised $31m in Series A funding led by Nava Ventures.
- The company monitors coolant, oil, and water for contamination, wear patterns, and fluid degradation.
- Liquid-cooled AI data centres are creating new maintenance requirements around fluid condition, chemistry, and failure detection.
Omen AI has raised $31m in Series A funding to expand continuous fluid monitoring for AI data centres and other mission-critical machines, bringing coolant chemistry closer to the uptime conversation.
The round was led by Nava Ventures and takes Omen’s total funding to $41.5m. The company says its technology attaches to fluid systems, including coolant, oil, and water, and tracks metal content, contamination, wear patterns, and degradation in real time.
Omen’s own site describes real-time coolant monitoring for data centres that detects bio-contaminants, metal wear, and fluid degradation. The company says its customers own and operate 10GW to 14GW of data centre capacity and manage about $200bn of data centre assets.
Liquid cooling changes maintenance
AI data centres are moving towards liquid cooling because rack densities are rising beyond the comfortable limits of conventional air-cooled designs. That transition changes the operational surface of the facility. Coolant distribution units, manifolds, cold plates, rear-door heat exchangers, pumps, seals, filters, and pipework all create new maintenance tasks and failure modes.
Fluid condition becomes part of reliability engineering once liquid moves closer to expensive compute equipment. Biological growth, corrosion, particulate contamination, incompatible fluids, seal degradation, pump wear, and blocked cold plates can all erode performance. Some issues develop gradually and may remain hidden until thermal alarms, equipment throttling, or service interventions expose them.
Traditional fluid testing often relies on periodic sampling and laboratory analysis. That approach can work, but it leaves a delay between condition change and operational response. Continuous monitoring is designed to reduce that gap by turning fluid condition into live telemetry.
That data only has value if it can be trusted and acted on. Operators will need to validate sensor accuracy, false-positive rates, calibration, installation impact, integration with building management and DCIM platforms, cyber security, and maintenance workflows. A tool that creates more alarms without clear intervention paths will not reduce risk.
AI racks raise the cost of small failures
The commercial logic is stronger in GPU-dense environments. High-density racks carry greater capital cost, and downtime can affect committed cloud capacity, model-training schedules, service levels, and revenue. A cooling fault that would once have been a maintenance issue can become a capacity event if it affects a cluster reserved for AI workloads.
Liquid cooling also changes staffing and procedures. Facility teams need knowledge of fluids, cleanliness, commissioning, leak response, flushing, inhibitor management, and vendor warranty conditions. Retrofitting liquid systems into older halls adds further complexity because pipe routes, service access, containment, and monitoring may not have been designed into the original facility.
Omen’s funding reflects a broader secondary market forming around high-density cooling. Beyond CDUs and cold plates, operators need leak detection, fluid monitoring, commissioning services, maintenance training, spares strategy, and standard operating procedures. The ecosystem around liquid cooling is becoming as important as the cooling hardware itself.
Europe’s data centre market is likely to see more of this operational detail as AI-ready retrofits and dedicated high-density halls move forward. Power availability may dominate site selection, but ongoing performance will depend on whether cooling loops remain clean, stable, and maintainable.
Omen’s raise is therefore not only a startup finance story. It marks the arrival of fluid chemistry as a measurable part of data centre operations. As heat moves into liquid, uptime engineering moves with it.
