NUMECA International
HPC Development Engineer
NUMECA International, San Francisco, California, United States, 94199
Main Responsibilities
As a Hardware / Software Interaction Specialist in the HPC Development Group. This position is based in the San Francisco area. Your main responsibilities will be
Optimize NUMECA’s legacy and next‑generation software suites for efficient execution on massively parallel, heterogeneous architectures. Perform in‑depth profiling and tracing studies to identify and resolve performance bottlenecks. Implement new functionality for improved strong scalability and a streamlined end‑user experience. Port NUMECA software to new compilers and platforms. Coordinate with other members of the HPC development group on bugfixes, research projects, and customer support. To be able to assume this role, the minimum qualifications are
BS in computer science, computer engineering, or a related field. Proficiency with C / C++ and a Linux development environment. Demonstrated ability to identify and resolve performance bottlenecks in large scientific applications. Experience with one or more of the following
Distributed memory parallel programming with MPI Shared memory parallel programming with OpenMP GPU Programming with OpenACC, CUDA, or OpenCL The additional qualifications are
Knowledge of compiler optimization techniques. Proficiency with Linux build tools such as make, cmake, and configure. Experience with profiling and tracing tools such as Intel VTune, gprof, and HPCToolKit. Experience optimizing code for modern CPU features such as advanced vectorization extensions. You should be highly motivated and dynamic, have good communication and analytical skills, be a stress‑resistant problem solver, be a team player, able to meet the highest quality standards, and ideally have a passion for programming.
#J-18808-Ljbffr
As a Hardware / Software Interaction Specialist in the HPC Development Group. This position is based in the San Francisco area. Your main responsibilities will be
Optimize NUMECA’s legacy and next‑generation software suites for efficient execution on massively parallel, heterogeneous architectures. Perform in‑depth profiling and tracing studies to identify and resolve performance bottlenecks. Implement new functionality for improved strong scalability and a streamlined end‑user experience. Port NUMECA software to new compilers and platforms. Coordinate with other members of the HPC development group on bugfixes, research projects, and customer support. To be able to assume this role, the minimum qualifications are
BS in computer science, computer engineering, or a related field. Proficiency with C / C++ and a Linux development environment. Demonstrated ability to identify and resolve performance bottlenecks in large scientific applications. Experience with one or more of the following
Distributed memory parallel programming with MPI Shared memory parallel programming with OpenMP GPU Programming with OpenACC, CUDA, or OpenCL The additional qualifications are
Knowledge of compiler optimization techniques. Proficiency with Linux build tools such as make, cmake, and configure. Experience with profiling and tracing tools such as Intel VTune, gprof, and HPCToolKit. Experience optimizing code for modern CPU features such as advanced vectorization extensions. You should be highly motivated and dynamic, have good communication and analytical skills, be a stress‑resistant problem solver, be a team player, able to meet the highest quality standards, and ideally have a passion for programming.
#J-18808-Ljbffr