Vincent Arel-bundock
Associate Professor
Montreal, Quebec, Canada
Summary
🤩
RockstarVincent Arel-bundock is an associate professor and data-focused software developer based in Montreal with 17 years of experience at the intersection of political science and reproducible data analysis. He contributes to widely used open-source tooling in R and Python—most notably improving tidiers in the popular tidymodels/broom package and enhancing model-performance diagnostics in easystats/performance. His work spans backend development, dataset curation (maintaining an Rdatasets archive), and API-driven data integration (World Bank data work for pandas), reflecting strong applied stats and data-engineering skills. He combines academic rigor with practical engineering, shipping CRAN-ready packages like modelsummary and contributing clear examples and diagnostics to statsmodels. Colleagues value his ability to translate complex econometric outputs into tidy, user-friendly formats that accelerate research workflows. An unassuming but prolific contributor, he often fixes edge-case model metrics and documentation details that substantially improve reproducibility for researchers.
16 years of coding experience
Github Skills (28)
apim10
python10
r10
package-development10
pandas10
statistics10
tidy10
regression-models10
latex10
api10
tidyr10
modeling10
documentation10
data-analysis10
data-science9
Programming languages (17)
Github contributions (5)
vincentarelbundock/modelsummary
Jan 2019 - Jan 2023
Beautiful and customizable model summaries in R.
Role in this project:
Backend Developer
Backend DeveloperContributions:14 releases, 9 reviews, 1711 commits in 4 years
Contributions summary:Vincent's commits focused on enhancing the modelsummary package's core functionalities. The commits include new example, and adjustments for the release of a new CRAN release. The user also worked on debugging and resolving issues related to the 'gof_map' and its handling of omitted goodness-of-fit (GOF) statistics. Additional arguments to `tidy()` method were implemented for greater control and flexibility, and support was added for `fixest` package.
r-packagestatisticshypothesis-testingbookdownregression-models
tidymodels/broom
Mar 2019 - Mar 2022
Convert statistical analysis objects from R into tidy format
Role in this project:Data Scientist
Data ScientistContributions:22 commits, 30 PRs, 68 comments in 3 years 1 month
Contributions summary:Vincent primarily contributed to the `broom` package, which focuses on converting statistical analysis objects into a tidy format. Their work involved adding and modifying tidiers for various statistical models, including those from the `ordinal`, `survival`, `MASS`, `mclust`, `tseries`, `car`, `Rchoice`, `cmprsk`, `mgcv`, `nnet`, `survey`, and `fixest` packages. They implemented new tidiers and incorporated confidence interval support where appropriate, improving the package's ability to represent model outputs in a consistent and user-friendly manner.
r-packagestatisticshypothesis-testingregression-modelstidymodels
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.