We’re finally getting a taste of proper summer weather, and it looks like it might stay that way for a while. So there’s a good excuse to spend some evenings outside, maybe even looking up at the night sky.
Even if we forget to do that, others are keeping watch for us. The Vera C. Rubin observatory has just begun operations, capturing breathtaking pictures of the southern hemisphere sky. Its hardware design is pushing the limits in several areas such as mirror engineering and camera sensitivity, but what might catch our community’s attention is the data. The observatory will generate around 20 terabytes of data every night for ten years, ending up with a dataset of around 60 petabytes. That is a lot of information to manage, and it will certainly keep a fair number of RSEs busy for years to come.
The observatory’s computing platform will provide browser-based tools for data discovery and visualisation, a JupyterLab interface for analysis, and remote APIs for programmatic access to international collaborators (including Imperial). These are familiar topics for us, and a good reminder of how RSE skills continue to shape modern science.
Let’s see what events, updates and resources are lined up for this month. Enjoy.
The STEP-UP RSLondon Conference will be held on Monday 7th July at One Birdcage Walk, Westminster. Full details are available on the conference web page. Attendance is free, and registration has been extended until Tuesday 1st July 2025.
rOpenSci’s July Social Coworking + Office Hours is scheduled for Tuesday 1st July 2025, 16:00-17:00 UTC. The theme of the month is “Research Software Engineering and R” with community host Saranjeet Kaur Bhogal. It is online and free to attend. More details are available here.
The NL-RSE meetup, titled “AI Accelerators for Scientific Applications”, will take place on Monday 7th July 2025 at the SURF office in Utrecht. The event will focus on how accelerators such as GPUs, TPUs and FPGAs are transforming scientific software. This in-person meetup is free and open to all. Details and registration are available on the event’s web page.
On Tuesday 8th July 2025 at 11:00 CEST, Judith Hartstein from the German Centre for Higher Education Research and Science Studies (DZHW) will join the HiRSE Seminar Series to present “Research practices in a digitalised world: Research software, mundane software and programming languages in research”. The Zoom link will be shared via various platforms, including the UK RSE Slack.
Based on the success of our previous Hackathon, the 2nd AI Hackathon will take place on Wednesday 9th July 2025 (09:30–17:00 BST) in RSM 2.42 at Imperial College London, bringing researchers and engineers together to prototype AI-driven tools. Lunch is provided. Please bring your laptop and any datasets or code you plan to use; standard AI API keys and starter code will be supplied, and extra resources (e.g. cloud access) can be arranged if needed. To participate, sign up to join or attend via this form – remote attendance is also possible. Feel free to forward this invitation to interested colleagues or groups.
Do you know Galaxy? Do you know what it can do for you and your research? In the seminar “Galaxy: the research data powerhouse:from analysis to FAIR management” we will introduce Galaxy, an open-source platform that empowers researchers across disciplines to analyse, manage, and share data without requiring programming expertise. It is widely adopted across life sciences, environmental research, astrophysics, and beyond. The event, followed by Q&A, will take place on Thursday 10th July 2025 at 13:00 BST online only (registration mandatory).
The third session of Imperial College London’s “Research Software Conversation Series”, titled “Research software documentation”, will take place on Tuesday 15th July 2025, 13:00–14:30 BST. The series, which runs monthly from May to July 2025, is free, hybrid, and open to anyone interested in research and research software.
Registration is now open for RSECon25, taking place at the University of Warwick, Coventry, from 9th to 11th September 2025. You can attend in person or join via hybrid format, with registration open until 31st July 2025. This year’s conference will focus on two interconnected themes: RSE and Research Excellence, and RSE as Digital Research Infrastructure.
The NES Col-Lab Retreat will be held in Schoorl, the Netherlands, from 27th to 29th August 2025. It will bring together researchers, data stewards, software engineers and support staff with an interest in the Natural and Engineering Sciences. Full details, including pricing and registration, are available on the event’s website.
There is an interesting satellite event – Back to Fortran Future 2 – organised as part of RSECon25. It will take place in person only on Monday, 8th September 2025, and there is no charge to attend. The workshop will explore practical ways the Fortran-based scientific community in the UK and beyond can enhance the quality of Fortran scientific software and improve software engineering practices. Registration is now open.
The 10th Open Data Camp is scheduled for 27th–28th September 2025 at the University of Edinburgh Business School. Organisers are currently seeking volunteers, sponsors, and community supporters. If you are interested, please get in touch at ODCampuk@gmail.com.
This month, in our series highlighting members of the Imperial community helping to support research computing, we hear from Francois van Schalkwyk:
Over two decades ago, I earned an MSci in Physics from Imperial College and pursued doctoral research in High-Energy Physics, modelling the response of low-light photosensors for a neutrino-oscillation detector. Drawn to hardware innovation and consumer electronics, I then joined an early-stage wearable-tech startup, leading PCB layout and developing embedded firmware for prototype devices. I subsequently moved into 3D-printer electronics, overseeing both hardware schematics and real-time control software.
More recently, I spent several years in the blockchain industry designing and implementing cross-chain bridges for Ethereum, Cosmos and other chains. I authored Solidity smart contracts, built Go and Node.js microservices containerised with Docker, and delivered REST APIs for enterprise clients. I also set up CI/CD pipelines with automated tests and code coverage, and supported security audits to ensure resilience under heavy load.
In September 2021, I returned to Imperial on a short-term contract to help develop Pybryt - a Python library that delivers automated feedback to students learning how to program with Python. When that engagement concluded, I joined the Department of Earth Science and Engineering as a Senior Research Software Engineer in February 2022. In this broad-scope role, I blend software engineering with DevOps - most notably standing up a departmental GPU cluster to accelerate machine-learning workflows.
I’m also passionate about teaching and mentoring. I’ve tutored numerous students and led hands-on workshops on topics from embedded systems to software development on blockchain. With the rise of large language models, I’ve had the opportunity to shift my focus to design, build and integrate LLM-driven solutions to support administrators, students, and researchers, beginning with highly targeted, domain-specific chatbots. The potential to augment human efforts and scale capabilities beyond previous limits is truly exciting, and I look forward to collaborating with others to push this technology to new frontiers.
This month, our Research Software of the Month is Kaira:
Communication systems research has traditionally been divided between two distinct approaches. Classical researchers focus on established error correction techniques - LDPC codes, Polar codes, and other proven methods that form the backbone of modern networks. Meanwhile, neural network researchers are exploring deep learning approaches that fundamentally challenge conventional communication paradigms. Kaira provides a unified platform that brings these traditionally separate research communities together.
The toolkit tackles a problem every researcher recognizes - good ideas getting stuck in academic bubbles. When your lab develops a promising new approach, comparing it fairly against existing methods shouldn’t require rebuilding everything from scratch. Kaira gives you a common foundation where battle-tested LDPC codes work alongside deep joint source-channel coding (DeepJSCC) models, all built to the same quality standards.
What’s impressive about Kaira is how much ground it covers. Whether you need 5G-standard Polar codes for a baseline comparison or want to pit neural Wyner-Ziv coding against everything from basic JPEG to cutting-edge BPG compression, it’s all there. The PyTorch backbone means you can actually optimize entire communication pipelines end-to-end, training encoders, decoders, channel models, modulations, and evaluation metrics together in ways that just weren’t practical before.
But here’s what really matters: Kaira is designed for real research collaboration. With 90% test coverage and 100% documentation coverage, the toolkit ensures both reliability and accessibility. Comprehensive examples guide researchers through practical implementations, while standardized benchmarks mean results actually have meaning, and the testing is thorough enough that you can trust what you’re building on.
As the field moves toward smarter, more semantic-aware networks, having a solid, shared foundation like Kaira feels essential. Finally, researchers can focus on the ideas instead of fighting with incompatible tools.
In last month’s newsletter, members of the central RSE team shared their experiences participating in the Collaboration Workshop 2025 (CW25). This month, a report on the Software Sustainability Institute (SSI) blog - “Work on skills and competencies for digital research professionals at CW25” - highlights a session run at CW25 by members of the DIRECT framework team. The rapidly developing DIRECT framework project is undertaking work to define skills and competencies framework for technical professionals.
Health Data Research UK (HDR UK) has launched a survey to gather insights from the technical community on preferred terminology for health data science roles, with the aim of improving future communications.
Last Tuesday 17th of June, the Imperial College London Research Software Engineering community hosted a session of the “Research Software Conversation Series: Tools and techniques for modern research in different domains”. This event explored how research software is used for hardware control and automation and, in particular, the challenges and lessons learnt when doing so. This blog post from the RSE Team summarises the main discussion points and take away messages.
GPTs are widely used, but the underlying concepts can be difficult to grasp. The “Visualising Transformers and Attention” YouTube video offers an accessible visual explanation of how transformer models and attention mechanisms work.
“Code for Thought” continues its monthly podcast series on topics relevant to research software engineering. Notable recent episodes include “Open Research – at FOSDEM 2025”, discussing developments in open research showcased at the well-known open source conference, and “Parsl: a Python library for parallel programming – with Dan Katz and Ben Clifford”, featuring a close look at the Python Parsl library and its use in parallel programming for research computing.
Speaking of FOSDEM 2025, videos of individual talks are being added to their respective session pages as they become available. You can also download recordings directly from this web page, organised by room name.
Following the news about the new space observatory, it’s worth highlighting tools that support space science research more broadly. Spacepy is a Python package designed to simplify data analysis, modelling, and visualisation in the space sciences. Built on NumPy and Matplotlib, it enables researchers to produce publication-quality results with ease. The project promotes open, reproducible research by offering a free, modern alternative to proprietary tools traditionally used in the field.
The website for Plotnine, the data visualisation package for Python based on the grammar of graphics, has recently been updated. As outlined in The Posit Blog, the homepage now focuses on inspiration rather than installation, showcasing the breadth of visualisation possibilities. The gallery has been enhanced, and a comprehensive new user guide is available.
Python 3.7 introduced the -X importtime option to display the duration of each module import, helping to identify startup bottlenecks. The python-importtime-graph online tool visualises this profiling output as a treemap, which can be saved in vector or raster image formats.
Data preparation is a crucial step in any serious data processing workflow. The blog post “10 Python One-Liners That Will Boost Your Data Preparation Workflow” presents concise Python code, primarily using Pandas, to handle common pre-processing tasks efficiently.
The Imperial Research Software Community Slack workspace is a place for general community discussion as well as featuring channels for individuals interested in particular tools or topics. If you’re an OpenFOAM user, why not join the #OpenFOAM channel where regular code review sessions are announced (amongst other CFD-related discussions…). Users of the Nextflow workflow tool can find other Imperial Nextflow users in #nextflow. You can find other R developers in #r-users and there is the #DeepLearners channel for AI/ML-related questions and discussion. Take a look at the other available channels by clicking the “+” next to “Channels” in the Slack app and selecting “Browse channels”.
If you want to start your own group around a tool, programming language or topic not currently represented, feel free to create a new channel and advertise it in #general.
If you need support with your code, seek no more! The Central RSE Team, within the Research Computing Service is here to help. Have a look at the variety of ways the team can work with you:
All the documentation, tutorials and howtos for using Imperial’s HPC are available in the Imperial RCS User Guide.
Imperial’s Research Software Directory provides details of a range of research software and tools developed by groups and individuals at the College. If you’d like to see your software included in the directory, you can open a pull request in the GitHub repository or get in touch with the Research Software Community Committee.
Drop us a line with anything you’d like included in the newsletter, ideas about how it could be improved, or even offer to guest-edit a future edition! rse-committee@imperial.ac.uk.
If you’re reading this on the web and would like to receive the next newsletter directly to your inbox then please subscribe to our Research Software Community Mailing List.
This issue of the Research Software Community Newsletter was edited by Stefano Galvan. All previous newsletters are available in our online archive.