Smart Building Operations

CoreWeave used an AI agent to anticipate thermal spikes and adjust liquid-cooling settings in live data center environments.

Data requirements

• Rack-power data (tabular) — Used to predict thermal spikes and guide cooling adjustments.

AI methods and techniques

• Reinforcement learning — A self-learning agent uses feedback from the environment to adjust cooling controls.

Industries

Real-world evidence

1 documented case study on record.

Companies using this: CoreWeave.