Close
Welcome to the virtual recap of UCL's Festival of Digital Research, Innovation & Scholarship, hosted by ARC
In July 2025, we hosted the Festival of Digital Research, Innovation & Scholarship for a second year!
A heartfelt thank you to everyone who participated—whether by showcasing your work or attending on the day. This year’s festival was a tremendous success and has grown significantly in its second year, with registrations up by 50% and a noticeable increase in poster, presentation, and workshop submissions.
Our ARC colleagues led some brilliant workshops, giving hands-on introductions to programming languages, research tools, and key digital skills.
We hope you enjoyed the festival as much as we did, and we’re already looking forward to next year.
Until then, don’t forget to subscribe to our newsletter to stay up to date with news and events.
The Festival in numbers
Programme revisit
Link to Full Programme Details including: sessions, abstracts, prize eligibility and voting instructions (PDF)
Welcome Tea & Coffee 9:00- 9:30 | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| 9:30- 9:45 | Welcome and Introducing ARC (Professor James Hetherington and Professor Jonathan Cooper, UCL ARC) | ||||||||
| 9:45- 10:05 | Opening Keynote - Emily Richardson, Amazon AWS, A Digital Survival Story: From research code to research infrastructure | ||||||||
| 10:05-10:15 | SAFEHR/Steven Harris | ||||||||
| 10:15-10:25 | Developing the UK SKA Regional Centre/Louise Chisholm | ||||||||
| 10:25-10:55 | Posters and introducing the sessions | ||||||||
COFFEE BREAK from 10:55-11:15 | |||||||||
Digital Twins | AI and Machine Learning | Enabling Research and Education | |||||||
| 11:15-11:20 | Introducing the Session | Introducing the Session | Introducing the Session | ||||||
| 11:20-11:30 | Keynote: Professor Mike Batty, UCL CASA | LEXCI: Learned EXascale Computational Imaging | RSE Initiatives Update + Broader Work in Knowledge Management | ||||||
| 11:30-11:40 | Professor Mike Batty, UCL CASA, continued | Real-time rapid diagnostic test (RDT) analysis using synthetically trained deep learning (DL) models deployed to mobile devices | High Dimensional Statistical Inference for Cosmology | ||||||
| 11:40-11:50 | Towards cancer digital twins: understanding how somatic mutation shapes cancer evolution | Evolution of Clay Microstructure Under 1-D Consolidation | Leveraging mailing list archives for digital history research: A case study on mining the Humanist discussion group | ||||||
| 11:50-12:00 | An Hybrid Digital Twin Framework for Taylor Vortex Reactor Synthesis | Towards a Non-Standardised Architecture: Co-Authored design through gestures at 1:1 scale | Rethinking the Evaluation of Educational LLMs: A Case Study of KELLY in L2 English Oral Proficiency | ||||||
| 12:00-12:10 | Data Integration for Space-Aware Digital Twins of Hospital Operations | Learned harmonic mean estimation of the Bayesian evidence with normalizing flows | Heterogeneity ²; Why Complex Biology Needs Complex Data | ||||||
| 12:10-12:20 | Building of the future: using digital building twins for data-smart operation | Challenges in RAG Evaluation for Text Classification in Evidence Synthesis | There and Back Again: from reproducible research to open education (and back) | ||||||
| 12:20-12:30 | Modelling Edge Computing Dynamics for Smart Cities: A Comparative Study between Markov Chains and Bayesian Networks in Barcelona | Novel AI algorithms for quality control of bioengineered organs | s2x: Differentiable and Accelerated Spherical Transforms | ||||||
| 12:30-12:40 | Questions | Questions | Questions | ||||||
| LUNCH BUFFET 12:40-13:30 | |||||||||
| Digital Twins | Data for Impact | Visualisation and Virtual Environments | Creating Data for Research | Communities | ||||
| 13:30- 13:35 | Introducing the Session | Introducing the Session | Introducing the Session | Introducing the Session | Introducing the Session | ||||
| 13:35-13:45 | Learning from Points of Interest to Understand City Spaces | EPPI Reviewer: Automated screening using large language models to support living evidence pipelines | gtexr: A convenient R interface to the Genotype-Tissue Expression (GTEx) Portal API | Gift: Unlocking Full Potential of Labels in Distilled Dataset at Near-Zero Cost | Experiences from 5 years of community-building in biological imaging and image analysis | ||||
| 13:45- 13:55 | Enhancing Fire Service Efficiency through a Digital Twin and Reinforcement Learning: A Case Study of West Midlands, UK | Digital Frontiers: An Open-Source Urban Data Platform for Queen Elizabeth Olympic Park | Tailoring fit for purpose 3D city models through generalisation | Scrambled text: fine-tuning language models for OCR error correction using synthetic data | Building a Scientific Computing in Rust community | ||||
| 13:55- 14:05 | Towards a Better Agent-Based Traffic Simulation for British Roads | Enhancing Due Diligence and Transparency through Blockchainand Privacy-Preserving Technology | Visualising Embodiment in Virtual Reality Interactions: Storyboarding as an Analytical Too | 360PanT: Training-Free Text-Driven 360-Degree Panorama-to-Panorama Translation | ReadingMOF: Systematic Chemical Names as Descriptors for Metal Organic Frameworks | ||||
| 14:05- 14:15 | Towards a Digital Twin Framework for Integrated EV Charging and Infrastructure Management | FetalSense – Transforming Fetal Monitoring for a Safer Pregnancy Journey | Evaluating the relationship between contingency and expressiveness in sandbox videogames and interactive simulations | Mechanistic Simulators for Rapid Diagnostic Test Capture and Deep Learning Classification | Working with the Grand Challenge of Data Empowered Societies | ||||
| 14:15- 14:25 | A digital twin for wastewater-based surveillance for SARS-CoV-2 and other pathogens: challenges and opportunities | Modelling Fire Service Efficiency in West Midlands, UK | Digital stereotomy for dry masonry architecture through interactive computational tools | Enhancing Image Segmentation Precision in Cystic Fibrosis Research Using Deep Learning | From Lab to Launch: The evolving role of RSEs in driving sustainable and commercially viable digital solutions | ||||
| 14:25- 14:35 | Evaluating the Role of Geospatial Science in Digital Twins: a National Mapping Agency Perspective | Customer Profiling Based on Mobile Apps GPS Data - A Case Study on Westfield Shopping Malls | NatureNest: Co-Produced Visualization Platform Connecting Community Gardening with Family Wellbeing | Using Driver Simulator to Explore Driver Behaviours in Different Experience Levels | ROS2 Community Group | ||||
Closing Plenary 14:45-16:15 | |||||||||
| 14:45- 14:50 | Welcome (Dr Claire Ellul, ARC Community Lead) | ||||||||
| 14:50-15:00 | Introducing UCL Press | ||||||||
| 15:00- 15:10 | Introducing UCL's Community Groups | ||||||||
| 15:10- 15:20 | The UCL Grand Challenges | ||||||||
| 15:20- 15:45 | Closing Keynote - Niall Robinson, NVIDIA,, Pulling Research through to Impact using AI and Digital Twins | ||||||||
| 15:45-16:00 | Closing Keynote: Professor Geraint Rees, UCL Vice Provost (Research, Innovation and Global Engagement) | ||||||||
| 16:00- 16:15 | Prizes and closing | ||||||||
| EVENING DRINKS AND NIBBLES 16:00-17:00 |
|---|
Workshops
- Computational tools for probabilistic modelling and statistical inference
Held by: Matt Graham and Samuel Livingstone
Probabilistic models are used across a wide range of research domains and there is a plethora of computational tools for constructing such models, fitting them to data and making predictions using them. This workshop will aim to bring together researchers across UCL who both use and develop statistical software tools. It will combine demonstrations of software packages, talks on applications of statistical tools and a discussion session where participants will be invited to flag gaps they have found in the current statistical tool landscape and areas they would like to see new tools being developed.
- Python-tooling: transforming Python scripts into a Python package with good software engineering practices
Held by: Matt Graham, Paddy Roddy, Sam Cunliffe and Saransh Chopra
The Python programming language is widely used in digital research and scholarship, due to its ease of use, wide ecosystem of existing libraries and open-source licencing. In this workshop we will showcase a project `python-tooling` started by UCL's Advanced Research Computing Centre to help researchers convert their scripts into a full-fledged Python package, combining a [website](https://github-pages.arc.ucl.ac.uk/python-tooling) and a [cookiecutter template](https://github.com/cookiecutter/cookiecutter). The workshop will consist of a demostration of how to use the template to set up a new project followed by a change for participants to try out the tools themselves and get advice on how to use in their own projects.
- Introduction to hospital data & the OMOP Common Data Model in R
Held by: Sarah Keating and Andy South
This workshop will introduce participants to the OMOP Common Data Model, which is becoming the de-facto standard for sharing patient records. We will also introduce R, a widely used statistical analysis tool, and give some examples of using OMOP data within this tool.
- Narrating FAIR: recognising and rewarding FAIR practices and the use of narrative CVs
Held by: Samantha Ahern and Vicki Yorke-Edwards
An interactive workshop for those with line management rsponsibilities to:
- Recognise benefits of FAIR practices
- Recognise FAIR on Narrative CVs
- Construct an example Narrative CV for your area demonstrating FAIR
- Express ways forward in recognising and rewarding FAIR in your contex
- First contribution to open-source software workshop
Held by: Sam Cunliffe, David Pérez Suárez, David Stansby, Saransh
Many of the software and tools we use every day in our research are open-source community projects. Can you imagine doing a data science project without Pandas or R? Could you imagine making plots and visualisations without Matplotlib or Plotly? Could you imagine writing a paper without LaTeX or Overleaf?
These and many other cornerstones of research software are all open-source and community-driven. They rely on bug reports and fixes from a community of volunteers and researchers.
This workshop is a hands-on introduction to making contributions to open-source software. We'll take you through the steps of identifying a project you rely on, finding a "good first issue" and making your first contribution; be that a pull request, bug report, or simply an improvement to some documentation. Contributing to open-source software projects is a great way to give back to the scientific research software community, build your coding skills, and gain recognition.
We'll be coding together in small groups, and there will be experienced open-source contributors from UCL on hand to help guide you through. Want to come but no idea what to contribute to?! There'll be a curated list of "good first issue"s from projects we maintain or frequently contribute to so you can get stuck in!
- Conversational Interfaces for Research: Transforming Data Interaction with LLMs and MCP
Held by: Cristian Dinu
What if the future of software delivery didn’t involve a user interface at all? With the rise of large language models (LLMs), we’re entering a new era where software can be consumed through natural conversation, not clicks. In this walkthrough, I’ll introduce the Model Context Protocol (MCP)—a new way to expose real-time, typed, dynamic data to LLMs, enabling them to act as intelligent, conversational front-ends for research software.
I’ll demonstrate how to use MCP, but also how to build a simple MCP server that exposes live data and commands to an LLM, transforming a traditional API into a human-friendly interaction. Along the way, we’ll look at what problems this approach is suited for, how it compares to conventional apps, and what design patterns are emerging.
- Open the Vault: Publishing and Sharing Your Research Data
Held by: Nicholas Owen and Samantha Ahern
This workshop is designed for all researchers eager to enhance the transparency, reproducibility, and impact of their work by effectively publishing their research data and software. In an era where open science principles are increasingly vital, this practical session will demystify the processes, best practices, and considerations for sharing your valuable research outputs across all domains.