Phaidra

Phaidra reports deploying autonomous cooling agents at all four Merck West Point chiller buildings by April 2022, averaging autonomous operation 84% of the time by year-end, and adding campus-wide AI Conductor coordination in early 2023.

Phaidra deploys autonomous AI agents for AI factories with customers and partners including CoreWeave, Applied Digital, NVIDIA, AirTrunk and InfraPartners, with reported reductions in cooling cost and improvements in energy efficiency.

Phaidra deploys autonomous AI agents for AI factories with customers and partners including CoreWeave, Applied Digital, NVIDIA, AirTrunk and InfraPartners, with reported reductions in cooling cost and improvements in energy efficiency.

Phaidra distinguishes read-only Prism with no control-system write paths from agents restricted to single objectives, equipment classes and defined permissions; connectivity is customer-controlled outbound TLS without persistent remote access.

Phaidra combines continuous-learning control models, AI-factory telemetry, root-cause analysis and agents for liquid cooling, power flexibility and chiller staging across complex data-center systems.

Phaidra combines continuous-learning control models, AI-factory telemetry, root-cause analysis and agents for liquid cooling, power flexibility and chiller staging across complex data-center systems.

Phaidra's official news page reports closing over $50 million in Series B investment led by Collaborative Fund, with Index Ventures, NVIDIA, Helena and Sony Innovation Fund.

Phaidra is integrated into AI-factory operations and partnerships with large infrastructure operators and data-center developers, including deployments tied to NVIDIA Max-Q AI factories and InfraPartners.

Phaidra is integrated into AI-factory operations and partnerships with large infrastructure operators and data-center developers, including deployments tied to NVIDIA Max-Q AI factories and InfraPartners.

Phaidra describes using contextual Class 3 operational data to train simulators or digital twins that generate Class 4 synthetic failure-state data, preparing agents for thermal spikes and equipment failures before those conditions occur.

Phaidra work spans control systems, reinforcement learning, data-center operations, energy optimization, observability and agentic infrastructure orchestration, giving highly valuable but domain-specific systems-AI skills.

Phaidra work spans control systems, reinforcement learning, data-center operations, energy optimization, observability and agentic infrastructure orchestration, giving highly valuable but domain-specific systems-AI skills.