Dily Duan Yi Ong

I am a Physics PhD student at the Kavli Institute for Cosmology and the Cavendish Laboratory at the University of Cambridge, working under the supervision of Dr Will Handley. Prior to joining Cambridge, I completed my undergraduate and master's degrees at Imperial College London. My research lies at the intersection of cosmology, advanced Bayesian inference, and machine learning, with a specific focus on rigorous model comparison and tension quantification.

I am the lead developer of unimpeded, a public pip-installable Python library and data repository designed to democratise access to computationally expensive nested sampling chains. Through this work, I have generated a comprehensive database across eight cosmological models and 69 datasets, transforming analyses that previously required months of supercomputing time into seconds. Most recently, I applied this framework to the DESI DR2 data release to critically examine claims of evolving dark energy. I am also a contributing author of anesthetic, a Python package for processing nested sampling chains.

Currently, I am extending this infrastructure to build the next generation of cosmological emulators. By leveraging machine learning techniques such as normalising flows and density estimators, I aim to enable instant, nuisance-free likelihood evaluations for future surveys.