Data centers operate 24/7 and generate large amounts of heat from thousands of servers. To cool the system, water is widely used thanks to its low cost and high efficiency. A large-scale data center can consume the amount of water equivalent to a small city.
Therefore, any leakage that appears in pipes, tanks or cooling systems can damage equipment, disrupt service and affect user data. This is why preventing water leaks becomes an important task for modern digital infrastructure.
At TechCrunch Disrupt 2025, startup MayimFlow won the championship in the Built World category thanks to a technology solution to prevent water leaks from causing damage.
MayimFlow founder John Khazraee said that many data centers only detect leaks when incidents occur. This forces the system to stop operating and could cause millions of USD in damage.
MayimFlow's solution uses IoT sensors combined with machine learning models to process directly on the system. This technology can predict leakage risks and issue early warnings from 24-48 hours, helping the operating team process in a timely manner, avoiding operational disruptions.
Khazraee has more than 15 years of experience working in the field of technology infrastructure construction for IBM, Oracle and Microsoft. He and his team, including veteran experts such as Jim Wong and Ray Lok, are aiming to expand solutions to commercial buildings, hospitals, factories and utility units.
Founder MayimFlow said that water is becoming one of the major global problems. Early detection of leaks not only protects data infrastructure but also helps save resources and operating costs.
