CPU Performance Architect

Overview In this role you will lead microarchitectural analysis on silicon to identify stalls, memory bottlenecks, and inefficiencies, driving root-cause resolution. You will design and run advanced power and performance experiments on physical silicon to validate pre-silicon projections against real hardware across general-purpose and AI/ML workloads. You’ll build early power and performance models to explore architectural scenarios and enable hardware-software co-optimization. Collaborating with lead CPU architects, you help set realistic power and performance goals for Google's future silicon, contributing to impactful products at scale. Compensation / Benefitsbonus targetequitybenefits ResponsibilitiesLead deep-dive microarchitectural analysis to isolate stalls and bottlenecksDesign, build, and run power and performance experiments on physical siliconDevelop and maintain early-stage power and performance models for what-if scenariosPartner with CPU architects and cross-functional teams to define dynamic power and performance goalsContribute to hardware-software co-optimization and future silicon initiatives Key requirementsBachelor's degree in Electrical Engineering, Computer Engineering, Computer Science, or related field8 years of experience with performance analysis, bottleneck isolation, and workload characterizationExperience with C/C++ for architectural modeling and Python for data analysis and automationCollaborative cross-functional collaborationStrong analytical and problem-solving abilitiesEffective communication of complex findings to technical and non-technical stakeholdersC/C++ for architectural modelingPython for data analysis and automationMicroarchitectural analysis and workload characterization