Calls

This call invites members of the University of Oxford to apply for up to 3 months of professional Research Software Engineering (RSE) support at 0.5 FTE. The purpose of the support is to help researchers prepare, adapt, or optimise research software for effective use of ARC and BMRC high-performance computing (HPC) resources. 

In this context, we consider research software to be either a single software tool or a pipeline that may include others’ software. 

The scheme is intended to support computationally intensive research where targeted engineering input would significantly improve performance, scalability, sustainability, or readiness for HPC environments.  

Proposed work is expected to lead to more effective, efficient, or appropriate use of ARC and/or BMRC compute resources. 

Eligibility 

• Applicants must be members of the University of Oxford (academic staff, postdoctoral researchers, or postgraduate researchers with supervisor approval). 
• Projects must have a clear plan to use ARC and/or BMRC compute resources. 
• Software may be at any stage of maturity, from early prototype to established codebase. 

What Support Covers (In Scope) 

Successful applications are likely to fall into one or more of the following categories: 

1. Performance optimisation, e.g. 
   • Profiling and optimising existing code to reduce runtime or memory usage on HPC systems 
   • Improving I/O efficiency for large-scale data processing 
2. Parallelisation, e.g. 
   • Introducing or improving shared-memory (e.g. OpenMP) or distributed-memory (e.g. MPI) parallelism 
   • Improving scalability across nodes and cores 
3. Accelerator and architecture-aware development, e.g. 
   • Porting code to GPUs (e.g. CUDA, ROCm, HIP, SYCL) 
   • Adapting code to better exploit vectorisation or modern CPU architectures 
   • Refactoring code into more performant languages or frameworks 
4. HPC readiness for new projects, e.g. 
   • Designing software architecture and workflows that will scale on HPC when the project matures 
   • Selecting appropriate languages, libraries, and data formats for HPC use 
   • Establishing reproducible build and run environments 
5. Software sustainability and maintainability, e.g. 
   • Improving code structure, testing, documentation, and reproducibility where these directly support HPC use 
   • Refactoring legacy research code to make it usable and performant on ARC/BMRC 

What Support Does Not Cover 

The following is a non-exhaustive list of things generally not in scope for this call: 

• Purely methodological or theoretical research without a software engineering component 
• Routine user support or training in how to log into or submit jobs to ARC/BMRC 
• Long-term maintenance or feature development unrelated to HPC performance or scalability 
• Software development that does not target ARC or BMRC usage 
• Data analysis tasks that do not require HPC-scale computation 

How to Apply 

Fill out the form linked below. You will be asked for the following long-form information, which you may wish to pre-prepare: 

• Brief project summary (1000 characters) 
  Describe the research aims and why computation is required. If this is part of a larger funded project, please give details as well as funder details and grant numbers. 
• Brief description of the existing software or planned technical approach (1500 characters) 
  Include current limitations, scale, and dependencies where relevant. 
• Description of the requested RSE work (2000 characters) 
  Describe what you want the RSE to work on, what success would look like, and any key technical risks or uncertainties. 
• Why is professional software engineering support required? (1000 characters) 
  Explain why this work requires specialist expertise or dedicated engineering time. 
• What will be possible after this support that is not possible now? (1000 characters) 
  Focus on scientific, technical, or scale-related impact. 
• How will the software be maintained or supported after the completion of the RSE support? 
  Describe who will maintain the software, for how long, and how the RSE work will be sustained. 

Application Form

Assessment Criteria 

Applications will be assessed on a rolling basis by members of the Oxford RSE team, and will be judged against the following criteria:  

1. Benefit to ARC/BMRC 
   The proposed work will lead to more effective, efficient, or appropriate use of ARC and/or BMRC compute resources. 
2. Technical Feasibility 
   The scope of work is clearly defined and realistically achievable within (at most) 3 months at 0.5 FTE. 
3. Research Impact 
   The work is expected to enable significant new scientific capability, scale, or performance. 
4. Value of RSE Support 
   The requested outcomes require professional software engineering expertise and would not reasonably be achieved without dedicated RSE input. 
5. Post-Project Sustainability 
   There is a credible plan for maintaining, supporting, or appropriately retiring the software after the RSE engagement ends.