Physics Meetup - 23 June 2026
Building on the resurrection of the regular RSDN meetups, we are organizing the inaugural RSDN meetup for Physics.
Research Software development is a core component of research in Physics. Within the department, there are many research software engineers working on cutting-edge, AI-driven software and developing high-performance algorithms. This meetup is intended to bring together everyone involved in research software or interested in research software. The meetup would comprise three short talks in the morning, followed by a networking lunch and an afternoon discussion on streamlining RSE activities within the department.
Talks by RSEs supporting research projects within Physics
- Brian Rogers - What We Don't See: An end-to-end machine learning stack for astrophysics data
- Adrian Oeftiger - Topic to be confirmed.
- Kaustubh Rajwade- OxRSE Group: A brief introduction
Discussion on improving RSE interaction and streamlining RSE activities in the department
|
OxRSE and the Schmidt AI in Science Fellowship - 19 May 2026
This meeting had an AI-in-Science focus, starting with an hour of short talks from members of OxRSE about the work they do supporting the Schmidt AI in Science Fellowship, followed by a networking lunch for in-person participants, and finally an optional 1.5 hr workshop titled “Responsible Use of AI for Coding”.
Talks by RSEs supporting the Schmidt AI in Science Fellowship Programme:
- Oliver King - Tracking and Identifying Picasso Triggerfish in Videos
- Jack Leland - TarXiv, a web app serving aggregated astronomical transient data from multiple sources; and roughpy-jax, a JAX backend for the Python maths library—RoughPy—for doing calculations on rough paths.
- Timothy Poon - Topic to be confirmed.
Responsible Use of AI for Coding Workshop:
This workshop guides attendees on integrating artificial intelligence into software engineering workflows effectively and ethically. Participants will learn to navigate the balance between the opportunities AI offers for rapid prototyping and the significant risks it presents, such as security vulnerabilities, hallucinations, and "sycophantic" bad code. The emphasis is on the fact that while AI serves as an "enthusiastic but inexperienced intern," the ultimate responsibility for code correctness, security, and maintainability remains with the developer. If possible, please bring a laptop to this session.
|