ABOUT US

UCL is one of the world’s top universities based in the heart of London. It is a modern, outward-looking institution. UCL’s staff and former students have included 32 Nobel prize winners. It is a truly international community: more than one-third of our student body – more than 35,000 strong – come from 150 countries and nearly one-third of staff are from outside the UK.
DiRAC is the integrated supercomputing facility for theoretical modelling and High-Performance Computing -based research in particle physics, astroparticle physics, astrophysics, cosmology, solar system &; planetary science and nuclear physics (PPAN; collectively STFC Frontier Science). The facility provides a variety of computer resources that match machine architecture to the algorithm design and requirements of the research problems to be solved. DiRAC comprises three computer services hosted at four Universities around the UK: Data Intensive Cambridge; Data Intensive Leicester; Memory Intensive Durham and Extreme Scaling Edinburgh.

THIS POSITION IS FUNDED FOR 20 MONTHS IN THE FIRST INSTANCE: 12 MONTHS AT 1.0FTE (01.08.25 - 31.07.26) PLUS 10 MONTHS AT 0.6FTE (01.08.26 - 31.03.27).

Candidates will need to upload a CV, cover letter and contact details of at least 2 referees.
The full Job Description and Person Specification can be found in the “Available Documents” section below. The Vision & Approach section of the bid (including the work-package chart) is also attached below.
Informal enquiries regarding the vacancy can be made to: Alastair Williams, DiRAC Project Coordinator, alastair.williams@ucl.ac.uk

DiRAC has secured UKRI funding to deliver a GPU Performance Portability Training (GPPT) programme to address the dearth of PPAN researchers with the skills to write optimized software for accelerators.
In the first year, the role holder will develop a series of self-paced training modules covering: the principles of GPU programming and GPU performance analysis, for presentation to users through our website. The source material for these modules will be a set of recorded lectures developed and delivered virtually by subject experts from within the community, under the GPPT programme in WP1 & WP2. The role holder will also work with experts who have already developed optimized codes to create a series of ~6 code deep-dives which demonstrate the efficient utilization of large-scale computing resources using real world-examples.
In the second year, the role holder will use the training material as the source for the development of the technical content for two five-day in-person training workshops (~25 people each) and manage a team of 5 expert (Research Software Engineer) mentors to deliver the workshops. The role holder will also work with industry & community experts to develop a simple ChatBot, trained on the developed material, to support students during self-paced learning.