Senior Performance Architect - Heterogeneous Workload Optimization
Core
Architecting and maintaining custom profiling frameworks to measure, analyze, and optimize the interaction between design graphs in system memory and high-throughput kernels on GPUs for EDA workloads.
Role type
Senior Systems Performance Engineer (Heterogeneous Workload Optimization)
Builds
Next-generation profiling infrastructure for EDA applications
Domain
Electronic Design Automation (EDA) / High-Performance Computing (HPC)
Deliverable
production ML models
Required skills
CUDA programming model, GPU profiling (Nsight Systems/Compute), system profiling (perf, eBPF, VTune, Valgrind), distributed compute (Slurm, LSF, Kubernetes), memory analysis, predictive modeling for hardware procurement
Preferred skills
None stated
Technologies
CUDA, Nsight Systems, Nsight Compute, perf, eBPF, VTune, Valgrind, Slurm, LSF, Kubernetes
Responsibilities
Architecting custom profiling frameworks for CPU and GPU environments; Conducting deep-dive benchmarking of EDA applications; Using GPU profilers to detect inefficiencies; Developing tools to monitor and attribute high-watermark memory usage; Developing predictive models for hardware procurement and cloud instance selection
Seniority
Senior, hands-on IC
