This role is for a Research Software Engineer within the High Performance Computing team in the Department of Physics, who will improve the performance of existing codes on large systems, implement more efficient algorithms, and liaise with researchers to identify their requirements. The work will involve performance benchmarking on prototype hardware to aid future UK HPC provision.
Responsibilities
- Improve performance of existing scientific codes on large HPC systems.
- Implement efficient algorithms and port codes to new architectures, including GPUs.
- Perform performance benchmarking on prototype hardware to inform future purchasing decisions.
- Liaise with researchers and other Research Software Engineers across the UK to translate scientific problems into efficient software.
- Work with parallel scientific codes that run on multiple servers using middleware such as MPI.
- Assist in the development of major scientific codes for cosmology, particle physics, and nuclear physics as part of DiRAC and ICC projects.
Qualifications
- Good working knowledge of C/C++ and Python.
- Some knowledge of Fortran is helpful.
- Experience with parallel computing and MPI in a large‑scale HPC environment.
- Understanding of performance optimisation, profiling and benchmarking techniques.
- Strong communication skills and ability to work with scientists and software engineers.
Experience & Skills
- Experience in HPC (e.g., Cosmological simulations, DiRAC, COSMA) is advantageous.
- Familiarity with high‑performance scientific codes and their optimisation.
- Background in cosmology, astrophysics, particle physics or nuclear physics is desirable.
Disability Confident
We are a Disability Confident employer and welcome applications from individuals who declare they have a disability.
#J-18808-Ljbffr