PhD position in automatic code generation for exascale finite-element solvers

Are you seeking an exciting PhD position in the area of high-performance scientific computing? Are you passionate about distributed computing, GPU programming, and numerical algorithms?  Do you want to help us develop a new generation of computational tools to democratize the access to massively parallel computing systems? Then, apply to join us at Vrije Universiteit Amsterdam.

We are looking for a motivated Ph.D. candidate to strengthen our team at the computer science department of VU Amsterdam under the supervision of Tiziano De Matteis and Francesc Verdugo. You will develop new methods to simplify the access to modern High-Performance Computing (HPC) systems in the context of finite-element based scientific simulations. You will research new source-to-source compilation technology able to translate mathematical definitions of scientific problems into efficient computer code that can run on modern computing systems. You will help us to answer key research questions: How to automatically generate efficient code for different computer architectures (CPUs and GPUs), problem types, and finite element formulations? How to escape the abstractions imposed by domain-specific languages used in code generation? How to extend these techniques beyond a single CPU/GPU to exploit the next generation of exascale supercomputers?  As a Ph.D. candidate, you will have the opportunity to present your findings at national and international conferences, contribute to scientific publications, and start building your scientific career. You also will develop a portfolio in the context of education by contributing to teaching in the department as a teaching assistant. This 4-year position is expected to result in a dissertation leading to a Ph.D. degree in Computer Science. 

For more details and how to apply: https://workingat.vu.nl/vacancies/phd-student-in-automatic-code-generation-for-exascale-finite-element-solvers-amsterdam-1059586