Senior Staff GPU Datacenter Performance Engineer
Core
Lead post-silicon performance characterization and optimization of AMD Datacenter GPU products for HPC and ML workloads across single-node to multi-node deployments.
Role type
Senior Staff IC GPU Datacenter Performance Engineer
Builds
Datacenter GPU products optimized for HPC and ML scale-up and scale-out
Domain
Semiconductor / High-Performance Computing / Machine Learning
Deliverable
production ML models
Required skills
Computer architecture, GPU system organization, HPC/ML workloads, performance analysis, Python, C/C++, shell scripting, root-cause analysis, technical leadership
Preferred skills
ML frameworks (TensorFlow/PyTorch), high-speed I/O (PCIe), cross-functional program leadership
Technologies
XGMI, Infinity Fabric, PCIe, TensorFlow, PyTorch
Responsibilities
Define performance-characterization strategies for HPC and ML workloads; Establish scalable workload-execution and data-collection frameworks; Influence product power, thermal, and performance definition; Drive performance attainment in scale-up and scale-out configurations; Analyze power-thermal-performance interactions; Identify and mitigate architectural/system-level bottlenecks; Lead prototyping for architectural/firmware/driver enhancements; Drive complex root-cause investigations; Mentor engineers on performance methodologies.
Seniority
Senior Staff, hands-on IC with technical leadership
