Omen AI adds liquid know-how in $16 million Aquatic Labs acquisition

Omen AI, a 2024 startup building real-time sensors for data center cooling fluids, is acquiring competitor Aquatic Labs in a $16 million cash-and-stock deal — a combination of two companies that had been chasing the same problem from different angles.

The context is scale. US data centers burn through 17 to 19 billion gallons of water a year cooling hot-running chips, a figure the Water Foundation expects to triple within five years. Today, checking whether that cooling loop is healthy means mailing samples to a lab and waiting days for results. Omen's sensors instead watch for elements in the water that signal failing valves or bacterial growth that could choke off flow; Aquatic Labs' device measures alkalinity. Merging both into a single unit gives what Aquatic CEO Allan Adams called a "much richer picture of the water and the fluid."

The founders are a study in contrast. Omen's Zach Laberge dropped out of college to build his first company, which ran fluid analysis on hydraulic construction machines. Adams is a trained physicist who spent much of his career on ocean sensing as an MIT professor and Woods Hole Oceanographic Institute researcher; he becomes the combined company's chief science officer. The two were introduced by a filtration-space investor.

The teams are working to integrate both technologies into one sensor they expect to have on the market early next year. Omen says its sensors will be monitoring chips drawing 250 to 300 MW of power by the end of the year — roughly $5 to $7 billion worth of hardware — with an installation goal covering the equivalent of 1 GW of chips by the end of 2027. The economic pitch is uptime: when fluid chemistry drifts, operators typically dump the water and halt operations rather than risk damaging expensive equipment, so longer fluid life and fewer cleanups are the sell.

What to watch is whether the integration actually delivers on schedule, and whether the combined sales motion converts the buildout skepticism now hitting hyperscaler stocks. Laberge argues the company sits downstream of the capex cycle — projects financed a year ago — and that even a halved buildout leaves an enormous market. Adams, meanwhile, wants the technology to travel beyond data centers, where water goes in and returns to municipal systems and must be measured at every step. The open question is how much of that adjacent demand materializes once a single sensor ships.