As AI drives higher demand for compute, data centers are packing GPUs more densely and increasingly adopting liquid cooling — with an attendant rise in bacterial contamination and clogged coolant loops. Typical coolant is a blend of water and an additive that inhibits bacterial growth; increasing the water fraction improves heat absorption but raises contamination risk. When bacteria or debris clog a loop, operators often flush the system, a process that can require shutting down a rack for five to six hours and potentially cost millions of dollars.
Omen AI offers a compact spectrometer that continuously monitors coolant chemistry and can detect bacterial growth before it becomes a system‑level failure. “You’re not risking huge amounts of downtime because you have no insight into what’s going on chemically,” said Zach Laberge, Omen’s CEO and founder, describing the value of immediate chemical visibility for operators.
Today Omen AI announced it raised a $31 million Series A round led by Nava Ventures. Other participants included CRV, Vanderbilt University, Mann+Hummel, Starhill Holdings and Hard Launch Capital, along with personal investments from executives at Bridgestone, GM, Johnson Controls and TensorWave.
Laberge is a serial young founder: he started his first company in 2020 at age 14, raising $3 million to place sensors on construction equipment before that startup shut down and he left high school. His parents supported his unconventional path; his mother previously served as Ontario’s Minister of Education. Laberge founded Omen in 2024 to focus on fluid systems as a key vector for reliable industrial machinery and building infrastructure.
The company’s goal was to replace slow sample‑taking and lab analysis with continuous, on‑site awareness. Besides bacterial detection, Omen’s device can identify traces of copper or chromium — signs of pump wear — and silicon that may indicate failing seals.
Omen’s early heavy‑equipment customers included Caterpillar dealerships; because Caterpillar also supplies gas turbines and generators for on‑premises data center power, those relationships exposed Omen to building‑side fluid systems and the data center market. “That was kind of the transition,” Laberge told TechCrunch. Around six months ago many dealerships began placing sensors on turbines and asked if Omen could address building‑side systems as well.
Omen now works with about a dozen data center customers while developing its product, including TensorWave, which is building an AI compute cloud on AMD chips. “The fluid running through these massive systems is a critical variable that most of the industry is flying blind on,” said Piotr Tomasik, president of TensorWave. “Omen [sees] the future of infrastructure exactly the way we do, better monitoring to optimally support compute customers.”
While many organizations still mail fluid samples to labs for analysis, Omen is not alone in pushing on‑site analytics: Pyxis, a water‑monitoring firm, launched a data center coolant monitoring product earlier this month.
Laberge attributes his approach’s feasibility to recent advances in optical hardware and signal‑processing software. “Hardware is just cheap enough that it makes sense to play at scale, and then signal processing lets us make more sense out of the noise,” he said.
Since its 2024 founding, Omen has raised $40 million in total funding. With the new $31 million Series A, the company aims to expand deployments across data center and industrial customers to reduce unplanned downtime and the high costs associated with coolant contamination and component wear.
Why it matters
- Growing AI compute densities increase cooling demands and push operators toward riskier, higher‑temperature operation to boost efficiency.
- Coolant chemistry directly affects reliability: microbial growth, metal wear and seal failure can all trigger costly shutdowns.
- Falling hardware costs and better signal processing make continuous, on‑site fluid analytics practical, enabling faster responses and less downtime.
Omen’s combination of compact optical sensing and software analytics targets a concrete operational blind spot as data centers scale up liquid cooling for denser AI workloads.
