Staff Systems Engineer, AI/ML
Core
Lead workload-driven architecture strategy for AI/ML systems, defining how to study, model, and optimize workloads across hardware and software boundaries to inform SoC architecture and HW/SW co-optimization.
Role type
Staff individual contributor systems engineer (AI/ML performance & architecture)
Builds
SoC architectures, memory subsystems, and AI/ML acceleration pipelines for current and next-generation semiconductor products
Domain
Semiconductor manufacturing, AI/ML systems, hardware architecture
Deliverable
production ML models
Required skills
Analytical performance modeling (roofline, bandwidth-latency), CPU/SoC architecture expertise, memory hierarchy optimization, AI/ML workload characterization, cross-functional technical leadership, mathematical reasoning for quantized/sparse models
Preferred skills
MLIR/IREE/TVM compiler infrastructure experience, model quantization/sparsity techniques, RISC-V architecture knowledge, Vector/Matrix extensions expertise
Technologies
MLIR, IREE, TVM, DDR/LPDDR, NPUs, DSPs, SIMD pipelines
Responsibilities
Define performance KPI frameworks for AI/ML workloads, lead architectural discussions between HW/SW teams, build analytical models to project system-level KPIs, present findings to senior engineering leadership
Seniority
Staff, high-impact individual contributor with organizational influence