HPC Computational Scientist

Position overview

KAUST is looking for a motivated HPC Computational Scientist to provide advanced technical support for internally developed software on high-performance computing platforms, including Shaheen III CPU and GPU partitions. The role focuses on enabling research and data-intensive work across computational science and engineering, big data analytics, and artificial intelligence/machine learning.

Core duties

• Support research enablement on KSL HPC systems across computational science, engineering, big data, AI/ML, and related areas.
• Assist with the installation, porting, tuning, and optimization of HPC applications and libraries so they run efficiently on CPU and GPU architectures.
• Build, maintain, update, and deliver user training, internal reference material, and external-facing documentation.
• Develop HPC applications and software tools needed to support research activities on cluster and supercomputer resources.
• Work directly with faculty, researchers, collaboration partners, and industry partners to help achieve project goals.
• Provide prompt user support through phone, in-person help, email, and ticketing systems.
• Maintain strong service quality while resolving user questions and technical issues.
• Support installation, porting, and tuning of scientific applications, tools, and the data analysis software stack.
• Stay current with new developments in technology, systems, and scientific computing through learning, training, conferences, and collaboration.
• Track emerging hardware and software trends and contribute to benchmarking activities for future lab purchases.
• Prepare and review system, testing, and software installation documentation.
• Deliver one-on-one or group training to users on relevant topics.
• Develop parallel software tools and libraries to support research projects across cluster systems and subsystems.
• Create software benchmarks to support periodic system upgrades.
• Contribute to scientific communication, outreach, and publication efforts where appropriate.

Required competencies

• Hands-on expertise supporting computational science, data analysis, or AI applications and libraries in HPC environments.
• Proven ability to assist research work in an HPC setting.
• Strong experience working collaboratively in research-driven teams.
• Programming knowledge in HPC-related languages and tools such as Fortran, C/C++, Python, MPI, OpenMP, CUDA, and OpenACC.
• Evidence of publications in computational science, engineering, or data analysis venues is preferred.
• Strong initiative, responsiveness, and teamwork skills.
• Excellent analytical thinking, troubleshooting, and decision-making ability.
• Creative and innovative approach to technical problem solving.
• Good organizational judgment with a broad systems perspective.
• Ability to deliver high-quality output under deadlines.
• Planning, coordination, and organizational capabilities.
• Knowledge of standard concepts, practices, and procedures in the field.
• Understanding of project management principles and practices.
• Ability to contribute effectively in teams and deliver results.
• Comfort handling several projects at once.
• Ability to work well in multicultural environments.
• Strong written and verbal English communication, including the ability to present reports orally and in writing.

Education and experience

A Ph.D. in computer science, computational science and engineering, or a closely related discipline is required. Candidates should have at least five years of experience supporting and using large-scale computing platforms and related subsystems, along with relevant background in computational science and engineering or a similar field. Experience with parallel computing and storage systems is expected. Prior research experience should include one or more of the following: performance modeling through simulation or analytical methods, tuning of HPC codes, I/O and data management, data compression, heterogeneous and emerging architectures, node-level programming with CUDA/OpenCL/OpenACC/OpenMP directives, or distributed programming models such as MPI and one-sided asynchronous approaches. Experience working at scale, such as on Top500-class systems, is desirable.

Additional information

This role requires close collaboration with users and research teams, with responsibility for both technical enablement and operational support across HPC systems.