Let's talk about science! My broader interests lie in using cosmological observables from CMB polarisation to galaxy clustering to probe fundamental physics and understand (to some extent) the nature of dark energy, dark matter, and inflation.

I am also a Euclid Consortium Member (the launch of Euclid here!), where I worked implementing different pipelines for testing deviations from the ΛCDM model.

A central theme of my research is the development of interpretable machine learning tools for cosmology, with an emphasis on transparency and explainability. I would like to put some emphasis on these two, as they are crucial when developing trustworthy ML pipelines. I apply methods such as SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) to uncover how machine learning models make decisions when classifying cosmological scenarios. This allows us to produce intuitive visualisations that highlight which features of the data drive model predictions, a result that not only enhances trust in machine learning applications but also provides an understanding that could be linked to the theory.