Principal Platform Power & Performance Engineer

Full Job Description

Own end-to-end power and performance characterization of next-generation Azure compute platforms, from silicon and server bring-up through workload-level validation. Analyze the interaction between CPU/GPU utilization, memory, PCIe/I/O, firmware, OS power management, thermals, cooling, and workload behavior. Identify opportunities to improve: Develop and evaluate dynamic power/performance-management strategies across idle, partially utilized, and fully loaded systems. Build telemetry and analysis methodologies that connect low-level platform behavior with workload performance and infrastructure-level efficiency. Diagnose complex cross-layer performance and power issues spanning hardware, firmware, kernel/OS, drivers, accelerators, and workloads. Provide technical leadership across multiple platform programs, define methodology and strategy, and drive ambiguous cross-organizational problems from investigation to deployment. Master's Degree in Electrical Engineering, Computer Engineering, Mechanical Engineering, or related field AND 7+ years technical engineering experience OR Bachelor's Degree in Electrical Engineering, Computer Engineering, Mechanical Engineering, or related field AND 8+ years technical engineering experience OR equivalent experience. These requirements include, but are not limited to, the following specialized security screening: 15+ years of relevant experience in server systems/platforms development and validation in computer architecture, systems engineering, platform performance, or power/performance engineering. Deep understanding of modern server architecture including: Ability to design controlled experiments, interpret noisy system-level data, establish correlations, and distinguish workload effects from platform bottlenecks. Experience debugging performance or power issues that cross multiple hardware/software layers. Large-scale cloud or hyperscale infrastructure AMD, Intel, Arm, NVIDIA, or other modern CPU/GPU platforms GPU and AI/ML workload characterization Power capping, DVFS, CPU/GPU power states, memory power management, or PCIe power management Server/rack power modeling and telemetry AI training or inference performance Performance-per-watt benchmarking Firmware or BMC-based platform controls Thermal and fan optimization Workload-aware or utilization-aware power management Capacity, rack-density, or datacenter power optimization