I am a Doctoral Researcher at the University of Edinburgh developing a Bayesian graphical modelling framework for real-time uncertainty quantification and evidence fusion in complex-contested-multi-domain environments. The work deploys swarms of sensor-equipped agents (aerial, underwater, and ground-based robotics) into contested regions and fuses the incoming data into a coherent picture of what is happening and what is likely to happen next, giving a commander a principled basis for decisions. The “contested” condition is treated formally rather than assumed away: the fusion relies on robust statistical estimation that remains reliable when an adversary corrupts a bounded fraction of the incoming evidence, framed in the language of Byzantine corruption and data poisoning and intended to extend to a broader class of adversarial threat models, and that holds under heavy-tailed, non-Gaussian sensor noise. The emphasis is on the fundamental limits of real-time detection under corruption, so that the assurance case rests on provable bounds rather than heuristics.
The research is funded by the UK’s Ministry of Defence and is motivated by modular digital twin architectures for autonomous systems in defence and security contexts, where interpretable, high-fidelity predictive analysis is critical. It is carried out in close collaboration with the Defence Science and Technology Laboratory (DSTL) and the Royal Air Force’s Rapid Capabilities Office.
I studied Engineering (bachelor's and master's) at the University of Cambridge, specialising in Transformational Machine Learning. I previously interned at the Zuse Institute Berlin, where I worked on research in explainable AI. I am also co-authoring a forthcoming book with a senior lecturer in Hindu Studies at the Cambridge Faculty of Divinity.
I have held a Senior Modelling Engineer remit at Network Rail, where I have been closely involved in shaping R&D initiatives that integrate ML and advanced analytical methods for infrastructure fault prediction. Through this work, I have engaged extensively with specialists from organisations including Deloitte, Cognizant, Capgemini, Faculty AI, and Arthur D. Little.