I
Senior Power and Performance Engineer
HybridFull-timeSystems Software DevelopmentSemiconductor Manufacturing
Join an AI hardware engineering team responsible for power and performance optimization at pod scale. The role owns telemetry, modeling, benchmarking, and capacity metrics for large-scale AI data-center systems. This is a hybrid position based in Israel, with locations in Haifa and Petah-Tikva.
Responsibilities
- Own pod-level power and performance metrics, including performance per watt, capacity, and utilization.
- Operate and define requirements for modeling and telemetry platforms, providing feedback on accuracy and coverage.
- Aggregate silicon and rack data to pod level, reconcile measured and modeled results, and manage power and performance budgets.
- Drive power telemetry, capping, and dynamic power management using RAPL, P-states, and DVFS.
- Build workload characterization and benchmarking pipelines using SPEC, MLPerf, and AI workloads.
- Identify bottlenecks and support pod capacity planning.
- Coordinate with facilities, electrical, and platform teams on power distribution, PUE targets, and peak-demand management.
- Work within a hybrid model that combines on-site and off-site work.
Requirements
- Bachelor’s, master’s, or PhD in electrical engineering, computer engineering, computer science, or a related technical field.
- At least 5 years of experience in power, performance, systems optimization, or a related area.
- Deep knowledge of CPU/GPU power management, performance counters, and Linux power frameworks.
- Proficiency in data analysis with Python and/or SQL.
Nice to have
- Experience with PUE or thermal modeling.
- Experience with pod-level capacity planning.
- Experience scaling AI clusters.