Research Associate, Imperial College London
Kaan is a Research Associate in the Department of Aeronautics at Imperial College London, specialising in high-performance computing and performance portability. His research explores how scientific software can run efficiently across different computing architectures. Within Imperial’s Turbulence Simulation Group, he works on x3d2, a flow solver based on graphics processing units (GPUs), developing methods for multi-GPU computation alongside validation benchmarks and automated testing
Kaan completed his PhD at the University of Bristol in 2025, investigating breadth-first search algorithms for multi-GPU and field-programmable gate array (FPGA) systems. His research at Bristol also included GPU acceleration of the NEMO ocean model for the UK Met Office’s next-generation modelling programme and contributions to reproducible benchmarking through the ExCALIBUR project. He has contributed to GPU programming training, including the OpenMP tutorial at SC24 and the organisation of a oneAPI workshop introducing students to the differences between GPU and FPGA coding approaches
Through the CoSeC Fellowship, Kaan aims to connect researchers facing shared challenges in adapting scientific software to modern computing platforms. He hopes to share practical approaches to GPU programming, performance portability and reproducible benchmarking, supporting knowledge exchange across disciplines and helping research communities make effective use of emerging hardware
Interesting fact: Kaan’s academic journey began with a degree in electrical and electronics engineering, before evolving into microelectronics and then computer science