Anthropic is a public benefit corporation dedicated to creating reliable and interpretable AI systems. The Data Center Engineer, Resource Efficiency will focus on optimizing power and resource allocation in AI infrastructure, ensuring efficient compute operations while maintaining availability commitments.
Responsibilities:
- Build models that forecast consumption across electrical and mechanical subsystems, informing capacity planning, energy procurement, oversubscription targets and risks, including statistical modeling of cluster utilization, workload profiles, and failure modes
- Design IT/OT interfaces that bridge compute orchestration with facility controls, enabling real-time telemetry across accelerator hardware, power distribution, cooling, and schedulers
- Build and operate load management systems that use power and cooling topology to enable load management and power/thermal-aware placement to maximize throughput while meeting SLOs
- Partner with data center providers to drive design optimizations and hold them accountable to SLA-grade performance standards, providing technical diligence on partner architectures
Requirements:
- Bachelor's degree in Electrical Engineering, Mechanical Engineering, Power Systems, Controls Engineering, or a related field
- 5+ years of experience in data center infrastructure or facility engineering
- Demonstrated experience with data center power distribution and cooling system architectures
- Experience building or operating software-based power management, load scheduling, or control systems
- Proficiency in Python or similar languages for statistical modeling, simulation, or automation of data center infrastructure optimizations
- Familiarity with SCADA, BMS, EPMS, or industrial control systems and associated protocols (Modbus, BACnet, SNMP)
- Track record of cross-functional collaboration across hardware, software, and facilities teams
- Master's or PhD in Controls, Power Systems, or related discipline and 3+ years of experience in data center infrastructure or facility engineering
- Experience with accelerator-class deployments and their power management interfaces
- Background in control theory, dynamical systems, or cyber-physical systems design
- Experience with energy storage, microgrid integration, demand response, or behind-the-meter generation
- Familiarity with reliability engineering methods
- Experience with SLA development, availability modeling, or service credit frameworks
- Exposure to ML/optimization techniques applied to infrastructure or energy systems