EPSRC Doctoral Prize Fellow from the University of Bristol. I am working on using Diffusion Models for Simulation-based Inference within Digital Twins. My PhD research focused on active learning and weak supervision techniques within machine learning, particularly improving performance in the face of noisy data and labels. My research has been motivated by the context of astronomical data, where labelled data is often scarce, noisy or expensive to obtain. My work and expertise have enabled me to be involved and consult in the morphology classification pipeline of the recently launched ESA Euclid telescope.Alongside my research, I have published interactive open-source software that facilitates active learning and allows researchers to utilise the techniques in any domain. During an internship at Imagination Technologies, I optimised deep learning operations by implementing novel convolution layers for highly sparse LiDAR data, reducing latency in self-driving applications, resulting in 5 patents. Through my work, I strive to develop innovative solutions that can reduce the workload of human experts and improve the reliability of machine learning models.