About Me

I'm a Scientific Software Engineer at Climate X, where I build and scale the geophysical and financial models that help institutions quantify climate risk. Day to day, that means working across catastrophe modelling, expected annual losses, and climate adaptation costs—wrangling terabytes of geospatial data and making sure our Slurm cluster doesn't catch fire.

Before this, I spent five years at Imperial College developing solvers for ocean and plasma simulations. My PhD focused on multigrid methods for compatible finite element discretisations—essentially, making large-scale physical simulations faster and more accurate. I contributed to Firedrake, an open-source finite element library, and I'm a co-author on its user manual.

I like working across abstraction boundaries. I've written preconditioners for the next generation of ocean models, built ML postprocessing pipelines at the Met Office, briefed the Government Chief Scientific Advisor on climate misinformation, and spent three months at Los Alamos shipping a fusion solver that became a publication. The common thread is starting with the problem and figuring out what actually solves it.

I also spent too many years tutoring A-Level maths, which taught me that explaining something clearly is often harder than doing it.

Experience

Climate X

Scientific Software Engineer · Jun 2024–Present

Productionising catastrophe loss models (EP curves, EALs, and downstream risk metrics) and creating models for adaptation finance. Led model validation for regulatory reporting, scaled pipelines processing terabytes of geospatial data, and introduced CI/CD across ~15 production repositories.

Imperial College London

Mathematics Researcher · Oct 2019–Present

PhD on iterative solvers for ocean and plasma models. Contributed algorithms and documentation to the Firedrake finite element library. Thesis: multigrid methods for compatible finite element discretisations.

Los Alamos National Laboratory

Graduate Researcher · Aug–Nov 2022

Implemented novel multigrid solvers for fusion simulations. Delivered publication-ready results in three months in an unfamiliar domain.

Met Office

Visiting Researcher · Jun–Aug 2021

Developed ML-based postprocessing for the MOGREPs ensemble forecast system. Recommendations were adopted into operational pipelines.

Government Office for Science

Policy Advisor · Feb–May 2022

Produced policy briefings on climate misinformation for the Government Chief Scientific Advisor, Sir Patrick Vallance.

Also: 6+ years teaching mathematics at undergraduate and A-Level.

Publications & Talks

Publications

  • Ham, D.A. et al. Firedrake User Manual, 1st edition, 2023.
  • Wimmer, G.A., Southworth, B.S., Gregory, T.J., Tang, X. A fast algebraic multigrid solver and accurate discretization for highly anisotropic heat flux I: open field lines, 2023.

Selected Talks

  • PyCon 2023: "Maths in Motion: Python-Powered Modelling for Real-World Phenomena"
  • PyCon 2023: "My Top 5 Least Favourite Data Visualisations"
  • Copper Mountain 2023: Novel solver techniques for anisotropic heat flux in plasma modelling
  • Warwick SIAM-IMA 2021 (invited): Multigrid preconditioners for ocean models

Beyond Work

  • UK Mathematics Trust: Mentoring Olympiad-level students and marking national competition papers
  • Andover Town Council: Elected councillor; Deputy Chair of Planning Committee
  • Conference speaking: PyCon, Copper Mountain, SIAM, Firedrake annual meetings