Staff Scientist (HPC Computational) en University of Rochester

Position Summary:

We are seeking a highly motivated and skilled HPC Computational Scientist to join our team. The successful candidate will be responsible for porting and optimizing scientific applications written in C and Fortran to run efficiently on GPU architectures, as well as providing expert support in compiling, running, debugging, and optimizing internally developed codes on high-performance computing (HPC) clusters.

This role also includes implementing and maintaining CI/CD pipelines, unit tests, and integration tests, as well as managing Spack-based software stacks in a multi-user HPC environment. The ideal candidate will have excellent communication skills, work effectively in a collaborative team setting, and possess a strong background in scientific computing.

Key Responsibilities:

• Port scientific codes written in C and Fortran to run efficiently on GPUs (e.g., via CUDA, OpenACC, OpenMP offload).
• Optimize and parallelize applications for performance on large-scale HPC clusters, including multi-GPU and multi-node systems.
• Assist internal teams with compiling, debugging, and tuning in-house codes on HPC systems.
• Develop and maintain CI/CD workflows to support software development and deployment.
• Implement robust unit testing and integration testing frameworks for scientific applications.
• Manage and support Spack environments in complex, multi-user HPC systems.
• Collaborate with researchers, developers, and system administrators to support and improve HPC workflows.
• Write and maintain documentation to support users and development processes.

Required Qualifications:

• U.S. Citizenship (required – must be able to work on-site).
• Doctorate (Ph.D.) in Computer Science, Physics, Applied Mathematics, Engineering, or a related field.
• Strong programming experience in C and Fortran.
• Hands-on experience porting and optimizing scientific codes for GPU architectures.
• Familiarity with CI/CD tools (e.g., GitLab CI, Jenkins) and modern testing practices.
• Experience with Spack in multi-user, HPC environments.
• Solid understanding of Linux, parallel programming (MPI, OpenMP, CUDA), and performance tuning.
• Excellent interpersonal and communication skills.
• Proven ability to work collaboratively in a team-oriented research or engineering environment.

Preferred Qualifications:

• Experience with performance analysis and profiling tools (e.g., NVIDIA Nsight, Intel VTune, TAU).
• Knowledge of containerization tools (e.g., Docker, Singularity).
• Background in numerical modeling, simulation, or domain-specific scientific computing.
• Experience using version control systems (e.g., Git) and collaborative development workflows.

• a cover letter; 
• a curriculum vitae and a list of publications; 
• at least two letters of recommendation

SALARY RANGE: $115,000 to $180,000

The referenced pay range represents the minimum and maximum compensation for this job. Individual annual salaries/hourly rates are set within the job's compensation range, and determined by considering factors including, but not limited to, market data, education, experience, qualifications, expertise of the individual, and internal equity considerations.