Cloud engineer at Development Seed · Earth observation data systems, from design to production · Former physical oceanographer
I design and run the pipelines that turn satellite archives into analysis-ready, cloud-native datasets, and the checks that keep them correct once they are in production. Most of that is Python, Kubernetes and cloud-native formats like Zarr and STAC, for organisations such as ESA and EUMETSAT.
Before that I spent about ten years (2010–2021) as a physical oceanographer, working with gliders and moorings in the Mediterranean and the subpolar North Atlantic, then learnt the software side at OSE Engineering and the UK Department for Work and Pensions.
The part of the job I keep coming back to is the step between raw observations and the people who use them. The ocean years left a few things other people still use:
- a Mediterranean mixed-layer climatology, merged from 13 hydrographic databases into 140,000+ quality-controlled profiles
- the UK-OSNAP glider datasets at BODC, and m_oceanglider, the MATLAB toolbox I wrote to process them
- the OSNAP mooring processing toolbox, whose adaptation from the RAPID-MOC toolbox I led, later taken over by SAMS for its mooring cruises
Along the way: 30+ peer-reviewed papers (Google Scholar, ORCID) and about 200 days at sea.
What I'm working on now, my talks and my blog posts are on lhoupert.fr (projects · talks · blog).
Outside work I keep a small homelab and explore the UK and Europe in our campervan with my family 🚐
Website · LinkedIn · GitLab · Stack Overflow · Email