We've all been affected by rail delay, arriving late to or even missing important events because of problems on the network. This issue is not just anecdotal; by the end of 2024, UK national passenger network performance had reached its lowest point in over a decade, with just 62.1% of services running on time and 5.1% cancelled (ORR Report, 2024).

Having spent the preceding 4 years working in smart transport, building data and AI systems for one of the UKs largest rail providers, it was clear to me that performance challenges have common causes in timetabling, system planning, and traffic management that cannot easily be solved within the operational constraints of industry - where financial limitations mean the focus and investment must be on keeping things running every day. With support from industry, I decided academia was the best place to develop new solutions to these problems.

To this end, I am currently working towards my PhD here at Royal Holloway, University of London; exploring applications of machine learning in smart transport with a particular focus on conformal methods.

My research interests are statistical modelling, algorithmic learning, and remote sensing - primarily as applied to transport.  However, I have also worked in a number of other interesting problem domains, including knowledge systems, operational research, and fluvial geomorphology.