Hardware Systems Engineer - Data Center HWE

About the role

We are seeking an experienced Server Hardware Engineer with deep expertise in CPU architectures, compute performance analysis, and cloud-scale hardware optimization. In this role, you will evaluate next-generation server platforms, assess workload requirements, and guide strategic decisions on processor selection across x86, ARM, and accelerator-enabled systems. You will partner closely with SRE, platform engineering, and capacity planning teams to ensure our Compute cloud infrastructure runs on the most efficient and cost-effective compute platforms.","responsibilities":"CPU Architecture & Performance Engineering: Evaluate and compare CPU architectures (Intel, AMD, ARM, emerging, accelerators) for cloud and containerized workloads. Conduct microbenchmarking and workload-specific benchmarking (DB, caching, ML inference, microservices, network-heavy workloads). Analyze CPU metrics including IPC, cache hierarchy behavior, NUMA topology, SMT/Hyperthreading efficiency, turbo behaviors, and power/performance curves. Build performance models to predict workload scaling, throughput, and cost efficiency across CPU generations. Server Hardware & Platform Analysis: Assess server platforms including memory configurations, PCIe topology, I/O bandwidth, NIC configurations, and accelerator integration (GPU, SmartNIC, DPU). Optimize server configurations for Kubernetes clusters, virtualized environments, and bare-metal workloads. Collaborate with OEMs and silicon vendors on roadmap evaluations and feature testing. Cloud & Kubernetes Workload Profiling: Profile and analyze workload patterns across large-scale distributed systems. Evaluate CPU/memory utilization trends to identify bottlenecks, over-provisioning, or scaling inefficiencies. Provide input into cluster sizing, autoscaling strategies, and capacity planning based on real workload telemetry. Build dashboards and telemetry pipelines to visualize compute efficiency across clusters. Procurement & Cost Optimization: Lead CPU and server platform recommendations for procurement cycles. Model TCO (total cost of ownership) including performance per watt, performance per dollar, and rack-level density. Provide guidance on the right mix of CPU SKUs, instance types, and server configurations for various cloud services.