About me

I am a final year PhD student in Astrophysics at the University of Oxford (Balliol College), supervised by Harry Desmond, Julien Devriendt, and Adrianne Slyz. My research focuses on the dynamics of the nearby Universe, using constrained cosmological simulations, Bayesian inference, and machine learning to connect observations with theoretical models. I am particularly interested in distance-ladder cosmology, reconstructions of the initial conditions of the Universe, large-scale cosmic flows, the galaxy–halo connection, and the interface between cosmology, gravitational-wave astronomy, and machine learning.

Beyond astrophysical applications, I am motivated by the development and use of advanced statistical methods. I work extensively with Hamiltonian Monte Carlo and related gradient-based samplers for scalable Bayesian inference, and I am interested in hierarchical and field-level models that exploit these techniques. On the machine learning side, I explore graph-based methods and geometric deep learning. More broadly, I am drawn to simulation-based inference, normalizing flows, and Gaussian processes as ways to capture uncertainties and extract robust information from complex datasets.

Most recently, I held a Pre-Doctoral Fellowship at the Flatiron Institute, working with Shy Genel and Lucia A. Perez on the application of cosmological rescaling to dark matter merger trees. In 2024, I undertook a four-month academic visit at the Institut d’Astrophysique de Paris, where I worked with Guilhem Lavaux on Bayesian field-level reconstructions of the local Universe. As part of my M.Sc. in Physics at the Ludwig Maximilian University of Munich, I worked with Miguel Zumalacárregui and Marius Oancea on strong-field lensing of gravitational waves. During my undergraduate degree at the University of Glasgow, I worked with John Veitch and Chris Messenger on gravitational-wave data analysis.

You can download a PDF version of my CV, which includes my full publication list. Alternatively, you can explore my publications directly via ADS.