How do we train HPC RSEs in the age of AI-assisted coding?
AI-assisted coding has vastly improved and agentic AI opens up new possibilities for refactoring and maintenance.While AI excels at common patterns, challenges remain with HPC languages and hardware-specific optimisations where model training data is sparse, along with incorporating the necessary domain knowledge to be able to validate software against the research context and goals.Do traditional pedagogical approaches still work? As we shift from syntax mastery to critical evaluation, what are the key competencies that RSEs need, and how do they get them? This talk presents the work of the CHARTED & DRIFT projects on understanding how HPC training is changing.
Authors
- Neil Chue Hong (ORCID: https://orcid.org/0000-0002-8876-7606)
Publication Details
- Journal
- Figshare
- Published
- 2026-11-18
- DOI
- https://doi.org/10.6084/m9.figshare.35052127
- Primary Topic
- Teaching and Learning Programming
- Type
- article
- Field-Weighted Citation Impact
- 0.00