HPC Optimization Engineer

POSITION SUMMARY

The Simons Foundation’s Flatiron Institute opened in 2016 and conducts scientific research in-house, supporting teams of top computational scientists. The Scientific Computing Core (SCC) is part of the Flatiron Institute and is led by Ian Fisk. This core develops and deploys the computing infrastructure—including new computational and statistical methods and storage and data-handling systems—necessary to support the research activities of Flatiron’s Centers for Computational Astrophysics, Biology, Mathematics, Neuroscience, and Quantum Physics. It also acts as a creator and disseminator of software tools for use by various scientific communities.

Primary Responsibilities

As a member of the Scientific Computing Core, the software engineer will be supporting a variety of open source projects across the Flatiron Institute, helping to make efficient use of various hardware accelerators, in particular GPUs.

The engineer will assist with:

• deploying and optimizing various machine learning tools and applications
• porting and optimizing HPC libraries and applications to accelerators
• gathering and disseminating information about unified programming models and tools for developing portable accelerator code
• preparing codes for deployment at leadership-class facilities with GPU-heavy architectures.
• Responsibilities also include related software engineering activities: source code control (using git), debugging, documentation, testing and continuous integration.

This engineer will report to the Deputy Director of the Scientific Computing Core. This is a full-time position based in our offices in New York City.

MINIMUM QUALIFICATIONS

Education:

• B.S., M.S. or Ph.D. in computer science or related technical discipline.

Experience:

• At least three+ years of professional experience with software development
• Proficiency in C, C++ and Python required, familiarity with Julia, MATLAB, R, etc., is desirable
• Parallel programming paradigms, especially MPI and OpenMP
• GPU programming (CUDA or similar)
• Optimization tools (monitors, profilers, nvidia-smi, nvtop, perf, htop, etc.)
• Scientific programming libraries (BLAS, LAPACK, FFT)
• Portable/unified programming models and techniques such as OpenCL, SYCL, OpenMP/GPU offloading
• Linux-based HPC clusters required, familiarity with SLURM is desirable
• Using and managing terabyte+ data sets

Related Skills & Other Requirements:

• Knowledge of general software engineering practices such as
  • Full Software Development Life Cycle, coding standards, code review and version control systems (e.g., Mercurial, Git).
• Ability and willingness to learn new programming languages and environments
• Technical and scientific curiosity
• Professional communication skills and a demonstrated commitment to customer service

COMPENSATION AND BENEFITS

• The full-time annual compensation range is $150,000 - $180,000, depending on experience.
• In addition to competitive salaries, the Simons Foundation provides employees with an outstanding benefits package.

The Simons Foundation’s Diversity Commitment

Many of the greatest ideas and discoveries come from a diverse mix of minds, backgrounds and experiences, and we are committed to cultivating an inclusive work environment. The Simons Foundation actively seeks a diverse applicant pool and encourages candidates of all backgrounds to apply. We provide equal opportunities to all employees and applicants for employment without regard to race, religion, color, age, sex, national origin, sexual orientation, gender identity, genetic disposition, neurodiversity, disability, veteran status, or any other protected category under federal, state and local law.