I am a Senior Teaching Fellow at Imperial College London. When problems are too complex for conventional methods, hand them to computers. That is both what I do as a computational physicist and what I teach my students to do.

My research focuses on computational magnetism. I develop open-source finite-element, finite-difference, and Monte Carlo simulation tools, including Finmag, Fidimag, and Ubermag, and then use them to discover new physics. I have a particular interest in exploring magnetic topological quasiparticles, such as skyrmions and Bloch points, as well as general magnetic phenomena at the nanoscale. I have also designed and participated in experiments and analysed large experimental datasets at the European X-Ray Free-Electron Laser Facility. I was part of the EPSRC Programme Grant on Skyrmionics and the OpenDreamKit Horizon 2020 project and have collaborated with and supported scientists across discipline boundaries through various projects. My wider interests include modelling and simulation, research software engineering, reproducibility, generative AI, higher education, and open science.

I teach and supervise MSc students in computational science, data science, and AI. I also work with academics and professionals across disciplines, helping them adopt best computational practices.

Before letting a computer solve a problem for us, we need to know how to translate the physics into something a computer understands. That translation is modelling, and running it is simulation. Together, they have made computational magnetism a third pillar of research, standing alongside theory and experiment. Simulations now routinely accompany experimental measurements in the literature, and are often what turns an observation into an explanation. Micromagnetic modelling is the workhorse behind much of this work. In a single lecture we cannot cover the field in any detail, but we can uncover the basics that participants can build on afterwards.

We start by replacing individual atomic spins with a continuous magnetisation field and then sampling it on a discretised domain. From there we assemble the total energy equation term by term: exchange, Zeeman, anisotropy, Dzyaloshinskii–Moriya, and demagnetisation. We will build an intuition for what each term wants the magnetisation to do, and see how their competition gives rise to the structures observed experimentally, such as domain walls, vortices, and skyrmions.

We then turn to dynamics and the Landau–Lifshitz–Gilbert equation, again asking what each of its terms is doing to the magnetisation. This leads to the different drivers used in practice — relaxing a system into equilibrium by minimising its energy, or following its evolution in time — and to the choices that determine whether a simulation can be trusted.

Finally, we will connect these ideas to Ubermag (https://ubermag.github.io/), so that participants leave not only better equipped to understand and judge micromagnetic results in publications but also ready to run their first simulation.