A career rarely unfolds in a straight line. Mine began in France, curved sharply north to Scotland, continued east to England, stretched all the way to Australia — and then found its way back to Bordeaux with a new purpose entirely.
In 2002, I enrolled at Robert Gordon University (RGU) in Aberdeen, Scotland, completing a BSc in Biological Sciences (2002–2004) followed by an MSc in Instrumental Analysis Sciences (2004–2005). Studying in Scotland was a deliberate choice: the research culture, the rigour of the programmes, and the proximity to world-class environmental science institutes made Aberdeen an exceptional place to train. Besides, I had a strong incentive to improve my English and discover new people and traditions.
For decades, the soil beneath our feet was treated as a “black box” in global climate and ecosystem models. We knew that soil microbes were responsible for critical biogeochemical cycles, yet their immense diversity and the complexity of their interactions meant they were often simplified into static parameters or ignored entirely as “functionally redundant”. My career as a researcher has been dedicated to prying open this black box, using emerging molecular tools and ecological theories to understand how microbial communities regulate Earth’s climate, respond to global change, and can even be harnessed for environmental remediation.
Not every data consultant starts in a spreadsheet. Some start in a field, pushing gas-sampling chambers into the ground at dawn, or in a molecular biology lab, running gel electrophoresis late into the evening. This is the story of how a career in research science became the foundation for a new chapter in data.
My scientific career began in earnest with a MSc thesis on Multiplex Terminal RFLP (M-TRFLP) — a molecular fingerprinting method for characterising soil microbial communities with high resolution. That early work, co-authored with Brajesh K. Singh and published in Applied and Environmental Microbiology, laid the methodological groundwork for years of subsequent research.
Greenhouse gas (GHG) flux measurements are notoriously noisy. Sensors drift, weather disrupts sampling, and the relationship between soil conditions and gas emissions is non-linear. This post walks through the analytical approach used in environmental GHG research projects, using data from the EucFACE experiment as a reference.
The primary gases of interest in soil flux studies are:
Fluxes are typically measured in µmol m⁻² s⁻¹ (CO₂) or nmol m⁻² s⁻¹ (CH₄, N₂O).
A data pipeline moves data from one or more sources through a series of transformations to a destination where it can be analysed or served. Getting this right from the start saves enormous debugging effort later.
Most pipelines follow the Extract → Transform → Load (ETL) pattern:
A variation, ELT, loads raw data first and transforms it inside the destination (common with cloud data warehouses like BigQuery).
Managing dependencies is one of the first challenges you face when working on multiple Python projects. Two tools dominate the ecosystem: Python’s built-in venv module and Conda. This post explains when to use each and shows the essential commands.
Each project may require different versions of the same library. Without isolation, installing a package for one project can break another. Virtual environments solve this by giving each project its own sandboxed Python installation.