Vacancy for HPC/cloud specialist

Will Furnass
10 September 2019 11:00

We recently hired two more Research Software Engineers (RSEs) (including Bob Turner, who started last week) but are already looking for another teammate!

  • Do you have experience of developing software in a lower-level language (e.g. C++) and a sound understanding of parallelisation technologies, HPC and/or cloud?
  • Do you want to work on a variety of research software/infrastructure development projects?
  • Would you also be keen for the opportunity to teach others about research computing / software development best practices?
  • Does the idea of collaborating with and supporting researchers so as to achieve their research objectives excite you?

If so, this new vacancy may be of interest.

Projects

Initially, you will work on at least two projects:

  1. Developing, testing, profiling and optimising high-performance parallelised software (MPI, OpenMP or CUDA) relating to computational methods for biomedical applications. Working with project partners there is the opportunity to deploy codes on international HPC infrastructure.

    This will also involve investigating how such software can be deployed on cloud computing infrastructure (e.g. AWS, Azure, GCP etc) to support research on the CompBioMed2 EU project.

    An example of the kind of application that needs further development, testing and deployment is pFIRE, which is a library for elastic image registration. This warps related images to achieve maximal correlation, which for MRI scans could compensate for deformation of the patient between images. The software was developed in modern C++ by a former member of the RSE team and uses PETSc to distribute matrix/vector data and operations between MPI ranks.

    pFire: mapped annotations

  2. Providing research computing support (software development and training/advice relating to software development, the use of HPC and data workflows) to the INSIGNEO institute for in-silico medicine.

    Quite a number of INSIGNEO’s researchers have non-trivial research computing needs due to them working with high-resolution medical images (e.g. MRI, CT) and sensor datasets, needing to apply expensive transformations to that data and needing to create and solve high-resolution finite element/finite difference

Teaching and other opportunities

There is also time allocated for working on and (possibly developing) other projects and training relating to high-performance computing and software engineering. For example, you could develop and deliver workshops on how to parallelise a workflow using OpenMP or joblib.

The post provides an excellent opportunity for research software skills development. You will actively support the wider community of research software developers and will be encouraged to represent the team in national/international activities. For example, current team members are/have been:

We are a supportive group that values further development of our members and we support each others through the diversity of our members’ skills and experiences.

Applying

Educated to PhD level (or be close to completion) in a computationally based field or having equivalent industry experience, you will have a track record of software development. For example, you might have:

  • developed/extended one or more applications/libraries at work or during your PhD
  • have managed those projects using version control,
  • have developed mechanisms for testing them
  • and have written documentation for them.

Applications are welcome from both software development generalists and from RSEs with interdisciplinary technology-focused skills which either complement or extend our existing portfolio e.g. data analytics (including statistical computing, machine learning and deep learning), numerical computing, HPC (including multi-core and GPUs), etc.

The post is open-ended (subject to funding).

Apply here!

The closing date is the 4th October. Please do get in touch via rse@sheffield.ac.uk if you want to discuss any aspect of the role before applying!

Contact Us

For queries relating to collaborating with the RSE team on projects: rse@sheffield.ac.uk

Information and access to JADE II and Bede.

Join our mailing list so as to be notified when we advertise talks and workshops by subscribing to this Google Group.

Queries regarding free research computing support/guidance should be raised via our Code clinic or directed to the University IT helpdesk.