Principal Platform Power and Performance Engineer

What you'll do

• Lead power and performance strategy for next-generation AI compute platforms to improve performance-per-watt and total cost of ownership.
• Provide technical leadership and mentorship to engineers while leading cross-functional initiatives to solve complex platform challenges.
• Partner with hardware, firmware, silicon, and software teams to define requirements and deliver optimized platform solutions.
• Drive architecture for workload characterization, power management features, frequency/voltage optimization, and memory subsystem efficiency.
• Optimize interactions between silicon, power delivery, thermal solutions, and operating systems to maximize reliability and scalability.
• Lead performance characterization, profiling, and benchmarking across Azure datacenter deployments using AI training and inference workloads.
• Influence future silicon and platform roadmaps through analysis of workload behavior and emerging AI infrastructure requirements.
• Communicate technical strategies and trade-offs to engineering leadership and executive stakeholders to align organizational goals.

What we're looking for

• A Master's degree in Engineering and 7+ years of experience or a Bachelor's degree and 8+ years of engineering experience is required.
• Candidates must have at least 8 years of experience in compute and/or AI systems platform design and development.
• At least 8 years of experience with SoC, CPU, GPU, or AI accelerator architectures including power management and thermal solutions is required.
• At least 8 years of experience collaborating across hardware, firmware, operating systems, silicon architecture, and validation teams is required.
• Ability to pass the Microsoft Cloud Background Check every two years is required.
• Experience with hyperscale cloud infrastructure, large-scale AI compute platforms, and datacenter operations is preferred.
• Experience with custom silicon, AI accelerators, and silicon bring-up from pre-silicon validation through production is preferred.
• Knowledge of server architecture, power delivery systems, liquid cooling technologies, and AI/ML workload characterization is preferred.