# Omen AI Acquires Aquatic Labs: Why Coolant Monitoring Matters for AI Data Centers

The $16 million acquisition adds another measurement to liquid cooling, but monitoring fluid and automatically treating it are different steps.

By Paras Madan | 2026-11-02


Omen AI announced its $16 million cash-and-stock acquisition of Aquatic Labs on November 2. The deal adds real-time alkalinity sensing to its data-center cooling technology, with Aquatic founder Allan Adams becoming chief science officer. The practical aim is to catch changes in coolant before they interrupt cooling.

This is an AI infrastructure story rather than a new model release. Expensive compute still depends on physical systems working reliably, and the condition of the fluid moving heat away from that hardware is part of that dependency.

## What Aquatic Labs adds to Omen AI

In its [acquisition announcement](https://www.prnewswire.com/news-releases/omen-ai-acquires-aquatic-labs-bringing-ocean-sensing-expertise-to-automate-data-center-cooling-302926594.html), Omen says Aquatic’s technology measures total alkalinity without adding chemicals for each test. Alkalinity describes a fluid’s ability to neutralize acid. It complements Omen’s elemental analysis and biological monitoring, helping assess corrosion and mineral-buildup risks.

The useful distinction is between seeing a problem after a scheduled sample and observing how conditions change between samples. A more complete set of measurements can help an operator distinguish an isolated reading from a developing maintenance issue. It does not, by itself, establish which intervention is safe.

## How Omen Sentry fits into a cooling system

Omen’s [product website](https://www.omen.ai/) describes Sentry as a rack-mounted system connected to the cooling loop through a parallel bypass. The listed measurements include pH, alkalinity, glycol concentration, conductivity, temperature and pressure. It supports Modbus TCP/IP integration with a facility’s building-management system.

That integration matters because another dashboard is not necessarily the outcome a data-center team needs. A useful monitoring system has to fit the existing operational process: who receives an alert, how it is checked, and which equipment can be serviced without disrupting a workload.

For a buyer, those workflow questions belong alongside sensor specifications. How often does a measurement arrive? How is the sensor calibrated? What happens when readings disagree with a laboratory test? A promising acquisition does not answer those questions for a particular installation.

## Monitoring is not autonomous coolant treatment

Omen expects products with the added monitoring capabilities to begin shipping by year-end. Automated treatment is a longer-term objective: connect analysis to equipment that adjusts coolant chemistry within approved limits, then measures the result.

**A measurement system and a control system carry different responsibilities.** Detecting drift can support a human decision. Changing the fluid automatically adds another requirement: the treatment must be appropriate for the coolant, the equipment and the operating conditions, with clear limits and a way to intervene.

The announcement does not provide quantified water savings or independently validated reductions in downtime. Those outcomes should be measured in deployment, not inferred from the acquisition price.

The next useful evidence will be whether continuous sensing helps operators make fewer disruptive interventions while keeping equipment within its required conditions. That is a more meaningful test than simply adding another sensor to the rack.

I’m Paras. I work at the intersection of AI, engineering and distribution.

I help technical founders explain complex infrastructure products through clear technical stories that developers and decision-makers can understand and share.

[Book a 30-minute call](https://cal.com/paras-madan/30min).