Current AI/ML practice often focuses on selecting a single “best” model while discarding non-optimal alternatives. This leads to missed opportunities because we lack effective ways to understand models' skill profiles, use them productively, or combine them meaningfully. Visualization can help address this challenge by leveraging the strengths of human visual perception to convey complex, high-volume information more efficiently than statistics or algorithms alone. It reduces cognitive load while preserving rich data context, enabling faster and more reliable reasoning. This EPSRC-funded project develops a visualization-enabled infrastructure and toolset that manages large pools of ML models and their performance profiles, supports the construction of effective model ensembles, and empowers decision makers to interpret model anomalies and reconcile conflicting predictions.
Co-I, Researcher
STFC, University of Oxford
Understanding protein mechanisms in cancer progression requires integrating data across imaging scales and modalities. This UKRI-funded project combines AI/ML with advanced imaging to analyze patient biobank samples, developing automated workflows to accelerate discovery of mechanisms behind cancer progression and treatment resistance.
Co-I
STFC
Nuclear fusion experiments generate massive experimental datasets that remain inaccessible to modern data science methods. In collaboration with UK Atomic Energy Authority, led the development of FAIR-MAST for the Mega Ampere Spherical Tokamak (MAST), designing infrastructure that makes fusion data adhere to FAIR principles and open-sources it for visual analytics and AI/ML—the first comprehensive data management system of its kind in fusion research.
Co-I
STFC
•••
UK infrastructure research increasingly depends on modelling and analysis at unprecedented scale. DAFNI is a national computing platform that enables researchers to run advanced simulations across transport, water, energy, and city-scale systems, generating insights that help make infrastructure more efficient, reliable, resilient, and affordable.
User Liaison
STFC
Led projects demonstrating the use of large language models (LLMs) at STFC for (a) grant proposal search and (b) semantic search, clustering, and topic modeling of scientific data from STFC's DataGateway.
Data Scientist
STFC, UKRI
During COVID-19, epidemiologists and modelling scientists required rapid, intuitive access to pandemic data and complex modelling outputs to make informed decisions. In response to the Royal Society's RAMP initiative, working with the University of Oxford and the Scottish Covid Response Consortium (SCRC), designed and developed RAMP-VIS, a visualization infrastructure that enabled SCRC scientists to efficiently analyse epidemiological data.
Postdoctoral Researcher
University of Oxford
The world's largest radio telescope, SKA, generates hundreds of TB/second of raw data, requiring real-time visualization at unprecedented scale. Designed and implemented a visual analytics system with low-latency data streaming architecture and web-based rendering pipelines for high-volume real-time data streams.
Postdoctoral Researcher
University of Oxford, SKA
Seismologists at the International Seismological Centre (ISC) previously relied on inefficient, paper-based workflows to review complex seismic events across multiple data types. Developed VBAS—an interactive visualization system that integrates diverse seismological data, including hypocentres, magnitudes, phase arrivals, travel-time curves, seismicity maps, station geometry, and more. VBAS provides a unified interface that helps ISC analysts detect patterns, identify anomalies, and perform data analysis far more efficiently. The system replaced the ISC's 30-year-old paper-based process.
Postdoctoral Researcher
University of Oxford, ISC
Development of a digital elicitation tool for climate resilience decision optimiser.
Research Software Engineer
STFC
Short project demonstrating capabilities of machine learning for crop disease prediction from hyperspectral data.
Researcher
STFC
Development of a comprehensive infrastructure for data, workflow, and provenance management and search for Horus Security Consultancy Ltd. UK
Postdoctoral Researcher
University of Oxford, Horus
Prototype enterprise search and visualization infrastructure developed as part of my DPhil research.
D.Phil. (Ph.D.) Research
University of Oxford
Software engineering for Oracle relational database management system.
Software Engineer
Oracle
Development of software for ABB's industrial automation controller.
Software Engineer
ABB