Richard is a Career Development Research Fellow at St John's College, Oxford, and a Hintze Prize Fellow in Cosmology. He studies the structure and expansion of the nearby Universe. He uses BORG to infer the Universe's initial conditions from galaxy clustering and peculiar velocities, and builds Bayesian models of the distance ladder to measure the Hubble constant.
He develops Bayesian methods for noisy measurements and incomplete samples. His hierarchical models account for measurement errors, selection effects and shared calibration uncertainties. He tests them on simulated data with known answers. He uses differentiable forward models, gradient-based samplers and simulation-based inference, and builds emulators to make expensive simulations practical for inference. His code runs on GPUs and computing clusters, handling models with millions of parameters. He works mainly in Python with JAX, NumPyro and BlackJAX.
He also works on galaxy formation, semi-analytic models, cosmological rescaling, forward modelling of the cosmic microwave background and machine learning for astrophysics. He is a member of the Aquila Consortium, Learning the Universe and CAMELS collaborations. He completed his DPhil at Oxford under the supervision of Harry Desmond, Julien Devriendt and Adrianne Slyz.