Performance Engineer, Deep Learning and HPC
Core
Generate benchmark data for Deep Learning and HPC frameworks/applications to drive marketing/sales collaterals and engineering studies.
Role type
Performance Engineer (Deep Learning & HPC)
Builds
Benchmark data reports and automated testing scripts for NVIDIA DL/HPC frameworks
Domain
High-Performance Computing (HPC) and Deep Learning (DL)
Deliverable
production ML models
Required skills
Python scripting, Linux system administration, software compilation from source, data analysis, debugging, containerization (Docker/Singularity)
Preferred skills
GPU-enabled DL frameworks (PyTorch, MXNet, TensorRT), GPU-enabled HPC applications (LAMPS, GROMACS, Amber, RTM), cloud GPU benchmarking (AWS, GCP, Azure), software compilers (GNU, Intel Composer, PGI), cluster orchestration
Technologies
Python, Linux (Ubuntu, CentOS), Docker, Singularity, PyTorch, MXNet, TensorRT, LAMPS, GROMACS, Amber, RTM, AWS, GCP, Azure, GNU, Intel Composer, PGI
Responsibilities
Plan and execute performance benchmarking across HPC and DL frameworks; Aggregate, analyze, and generate written and visual reports; Develop Python scripts to automate testing; Debug performance issues with engineering teams; Assist in developing tools for automated testing
Seniority
Mid-level (2+ years experience)
