Horizons
Alongside astrophysics, I am interested in how the same modelling and scientific computing approach applies in other areas of physics. This is not a change of identity so much as a reminder that the methods travel well.
Environmental physics
- Process-based models where parameter choices matter (and need to be tested systematically).
- Coupling physical sub-models with observational constraints and clear validation checks.
- Reproducible, well-documented workflows that make assumptions explicit.
Meteorology and atmosphere
- Model evaluation and comparison across ensembles and parameter sets.
- Practical data handling and pipeline building for gridded and time-series products.
- Clear technical communication: what a model does, what it assumes, and where it fails.
- Supervising a 2026 student project on AI hurricane-track prediction.
Methods I use
I am comfortable working from first principles and then building the software needed to test ideas quickly and reproducibly. My core tools are C and Python, with a preference for simple, transparent implementations that are easy to audit.
- Scientific computing: performance-aware code, numerical experimentation and careful testing.
- Model development: exploring sensitivity to assumptions through controlled parameter studies.
- Workflow design: automation, traceability and documentation so results can be reproduced.
If you are working in an applied area outside astrophysics, I am happy to discuss how these methods map to your domain.