About Me
Physicist, machine learning researcher, and educator working at the intersection of AI and fundamental science.
Background
Previous Affiliations
University of Cambridge
University College London
Education
PhD in Physics & Astronomy
University of Heidelberg, 2018
AI for Science
I lead the "AI for Science" theme within the Centre for AI and Skills at Royal Holloway. This role involves fostering collaboration across multiple departments — including Engineering, Mathematics, Physics, Earth Sciences, and Computer Science — to integrate AI methodologies into scientific research.
I am a member of the STFC Computing Advisory Panel, advising the UK government on national strategy for digital research infrastructure and AI in science.
Research Vision
My research places uncertainty quantification at the heart of scientific discovery. I develop machine learning and Bayesian methods that go beyond point estimates to provide rigorous, calibrated uncertainties — enabling robust scientific conclusions from complex data.
This vision unifies two seemingly disparate fields: cosmology, where I study the accelerated expansion of the Universe and test gravity on cosmic scales, and seismology, where I apply the same probabilistic techniques to understand how earthquakes rupture.
I have been awarded a Leverhulme Research Leadership Award to establish ECLIPSE (Environment for Computational Learning, Interdisciplinary Physics and Scientific Excellence), a new interdisciplinary research hub that will pioneer uncertainty-aware AI-driven scientific discovery across cosmology and seismology.
Euclid
I play a leading role in the scientific exploitation of ESA's Euclid satellite mission. I hold multiple leadership positions within the Euclid Consortium:
- Lead of the Key Project "Euclid cosmological constraints from combined photometric probes", delivering flagship results from Data Release 1
- Lead of the 3x2pt Work Package, responsible for headline cosmology results from Euclid weak lensing and galaxy clustering data
- Lead of the "Simulation-Based Modelling and Exploratory Techniques" Work Package, developing simulation-based and machine learning tools to maximise Euclid's scientific return
I am the creator of COSMOPOWER, a machine learning framework for accelerated cosmological inference that has been adopted by multiple international collaborations.
Grants
- PI, Leverhulme Research Leadership Award: £997k
- PI, ISSI International Team grant (Euclid DR1)
- PI, DiRAC allocation: 5.74M CPUh
- Co-I, EuroHPC Extreme Scale Access: 10M CPUh
- Co-I, Perren Fund for Astronomy: £87k
- Co-I, STFC Small Award (INTIME): £463k
- Co-I, ARCHER2 software dev (GLASS): £355k
Distinctions
- Euclid Consortium Builder (2025)
- Invited IDEA Scholar, Flatiron Institute (2025)
- Robert Boyd Award for Outstanding Scientific Achievement (2022)
- Alan Turing Institute Post-Doctoral Enrichment Award (2022)
Publications
Find my publications on Google Scholar, Inspire HEP and arXiv.