I have several years of experience designing and developing scientific software with a focus on solving differential equations and finite-element methods scaling from workstations to high-performance computing (HPC) environments. I have also designed embedded domain-specific languages (eDSLs) in Python to describe scientific problems at a high level and with high expressivity. Furthermore I have some (not professional) experience in full stack web application development.
Currently I work as a computational scientist at ECMWF, designing and developing scientific HPC software for processing large amounts of metereological forecast data.
Previously I was a PhD researcher in the Software Performance Optimisation Group at Imperial College London, delivering generic and efficient computational solvers for partial-differential equations through automated code synthesis from a high-level problem description down to a highly optimised parallel implementation specifically tailored to the target hardware. Read more about this on my research page or look at my publications.