Sebastian Kuckuk

Dr. Sebastian Kuckuk is a researcher and head of training at the Erlangen National High Performance Computing Center (NHR@FAU), Friedrich-Alexander-Universität Erlangen-Nürnberg. He is affiliated with the Department of Computer Science and contributes to the Chair of System Simulation. His work bridges research, training, and software development in high-performance computing.

Education:

• PhD in Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg (2019)

His research focuses on enhancing performance portability and programmer productivity using domain-specific languages, code generation, automatic parallelization, and GPU programming. These techniques are applied to develop massively parallel numerical solvers for computational fluid dynamics, particularly for the shallow water equations. He is a core developer of the ExaStencils framework, which enables automated generation of efficient multigrid solvers for structured and patch-structured grids.

Analysis of his recent publications (2020–2025) reveals a consistent focus on code generation, GPU acceleration, and solver optimization for fluid dynamics problems. Key themes include heterogeneous computing, block-structured grids, and adaptive methods. His work integrates advanced compiler techniques with numerical mathematics to improve scalability and performance on modern HPC architectures.

Scientific Recognition:

• NVIDIA Deep Learning Institute (DLI) University Ambassador
• Certified Instructor for DLI courses in GPU programming and CUDA

He actively contributes to teaching and training through courses such as Programming Techniques for Supercomputers and High-End Simulation in Practice. He conducts workshops and tutorials on GPU programming and performance optimization. While no formal students are listed, his mentoring role is evident through collaborative research and training activities. He has no recorded grants in the provided text, but his involvement in NHR and KONWIHR projects indicates active participation in funded HPC initiatives.

Laboratories and Projects:

• Lead developer of ExaStencils, a code generation framework for multigrid solvers
• Contributor to GHODDESS, a module for higher-order discretizations in shallow water modeling
• Active in NHR@FAU and KONWIHR projects focused on GPU computing and performance optimization