I am a PhD Student in Political Science at the University of Konstanz (Germany). My research lies at the intersection between Political Methodology (with a taste for textual data), Legislative Politics (small number of actors for a lot of texts) and Online Political Behavior (TikTok is purely professional!). In my dissertation, I develop new methodological tools to measure the ideological position of MPs and test spatial models of legislative politics.

Beyond academia, I am a compulsive #rstats programmer and a dedicated fooddy! I also spend part of my life in Canada sharpening my wild lumberjack skills: I can’t get enough of free space, wood chopping and mapple syrup.


Legislative Politics

My substantial interest lies in European legislative politics and my dissertation builds up new measurement strategy for the ideology of political actors. Using these new measures, I investigate how parliamentary leaders deal with conflict in their own party and how they limit the public costs of free-riders holding extrem views. Beyond ideology, I am generally interested in the role played by parliaments in the policy-making process.


As NLP thrives, I try to import its recent progress toward political science research. I have used both supervised (embedding, CNN-LSTM) and unsupervised (topic model) approach, although I definitely prefer the former for its intuitive and straight-forward validation. Lately, I have tried to follow up on the transformer revolution and currently work on a wrapper for the Huggingface library.

Data Management

A substantial portion of my work is dedicated to large-scaled data collection processes. I have scraped Facebook, Google, Twitter, Tiktok using among other a home-made hive of raspberry pies. In parallel I developed tidybrowse: a suite of Rpackages meant to ease web scraping.