Scientific Programmer (m/f/d) High-Performance Computing and Software Engineering

Descrizione

Testo originale dell'annuncio (in inglese)

The Max Planck Institute for Marine Microbiology is a research institute of the Max Planck Society. It was founded in 1992 in the State of Bremen and employs around 200 staff members. In close collaboration with numerous university and non-university research institutions, we explore the diversity and function of microorganisms in the ocean and their interactions with their environment. Scientists from all over the world, engineers, technicians and administrative staff together make an important contribution to global marine and environmental research. In the Department of Ocean Biosphere Dynamics, we are looking to fill a position as Scientific Programmer (m/f/d) High-Performance Computing and Software Engineering The department investigates the dynamics of the ocean carbon cycle, its variability and predictability, and its role in the Earth system through climate-carbon cycle feedbacks operating across timescales from days to millennia. A central objective is to improve the representation of marine carbon, nutrient cycling and microbial processes in Earth System Models while increasing spatial model resolution to capture key physical and biogeochemical dynamics in the ocean. To support this mission, we are seeking different Scientific Programmers (m/f/d) to drive the next generation of ocean biogeochemical modelling in Earth System Models. The successful candidates will contribute to the development, integration, and performance optimization of a state-of-the-art modelling framework, enabling increasingly sophisticated representations of marine biogeochemical processes and unprecedented high-resolution numerical experimentation. Together, they will help ensure that the modelling infrastructure combines scientific excellence with cutting-edge software engineering, delivering robust, efficient, and scalable performance on current and future exascale computing platforms. The successful candidate will help ensure that the modelling framework of the HAMburg Ocean Carbon Cycle model (HAMOCC) which is part of the ICON Earth system modeling framework remains technologically advanced and capable of efficiently exploiting emerging HPC architectures. Emphasis will be placed on heterogeneous computing and the optimal use of CPU and GPU resources to maximize model performance, scalability, and portability. A particular focus of the position will be a dedicated project to prepare, optimize, and benchmark the code for deployment on heterogeneous CPU/GPU systems, ensuring efficient performance and supporting the transition to next-generation HPC infrastructure. The role involves close interactions with researchers, research students and model developers to translate scientific advances into reliable and maintainable model code. The successful candidate will also have the opportunity to contribute to peer-reviewed publications and present research outcomes at scientific conferences. Description Key responsibilities: Develop and optimize model code of HAMOCC for modern supercomputing architectures Enable and improve heterogeneous execution across CPUs and GPUs Lead and contribute to software development, maintaining Git repositories, testing and debugging, code review, and Continuous Integration/Continuous Deployment (CI/CD) to ensure robust, maintainable, and reproducible code Implement performance analysis, profiling, and optimization strategies Support portability across current and future HPC platforms Collaborate with model developers to translate scientific requirements into efficient software solutions Requirements Desired qualifications: Degree in computer science, computational science, applied mathematics, or a related field Experience with numerical modelling and analysis and handling of large simulation datasets Strong knowledge of HPC programming (Fortran and C/C++) and performance optimization Knowledge of parallel programming paradigms such as MPI, OpenMP, OpenACC or equivalent technologies Experience with large-scale scientifi