Performance Engineer, Deep Learning and HPC

Santa Clara, CA

Full-time

Onsite

$116K/yr - $190K/yr

Entry, Mid Level

NVIDIA has been transforming computer graphics, PC gaming, and accelerated computing for more than 25 years. The Performance Engineer will focus on Deep Learning and High-Performance Computing applications, responsible for generating benchmark data and collaborating with multi-functional teams to improve automated testing processes.

Apply Now

Responsibilities

Plan and execute performance benchmarking across a wide range of HPC and DL frameworks and applications
Aggregate, analyze, and generate written and visual reports with the testing data for internal sales, marketing, SW, and HW teams
Develop Python scripts to automate the testing of DL & HPC-focused applications
Work with internal engineering team to debug performance issues
Learn to use the latest applications in the areas of advanced machine learning and HPC
Assist with the development of tools and processes that improve our ability to perform automated testing

Qualification

Required

B.S. in Computer Engineering, Computer Science or related technical field (or equivalent experience)
2+ years of experience
Excellent programming and debugging skills in a scripting language such as Python or Unix shell
Advanced knowledge using Linux based systems (Ubuntu and CentOS strongly preferred)
Proficient in compiling software from source code, including debugging errors encountered
Excellent English verbal and written interpersonal skills to improve collaboration with coworkers
Excellent data analysis skills and the ability to summarize findings in a written report
Familiarity using a container platform such as Docker or Singularity

Preferred

Experience using a GPU-enabled deep learning framework such as PyTorch, MXNet, or TensorRT
Experience using GPU-enabled HPC applications such as LAMPS, GROMACS, Amber, RTM, etc…
GPU/CPU benchmarking on cloud solutions from AWS, GCP, Azure
Software compilers such as GNU, Intel Composer, or PGI
Previous experience with computer clusters and cluster orchestration and scheduling tools