Simon Couch
Software Engineer at Posit PBC (née RStudio)
Chicago, Illinois, United States
Summary
🤩
Rockstar🎓
Top SchoolSimon Couch is a software engineer with eight years of experience specializing in statistical software, data pedagogy, and open-source development. Based in Chicago and currently focused on statistical tooling at Posit, he contributes to prominent tidymodels projects like broom and recipes, improving model integration, performance, and feature-engineering robustness. His work on broom includes nuanced handling of models from MASS and rstanarm and modernizing internals for better performance, reflecting both domain knowledge in applied statistics and practical engineering. Simon combines technical writing and community-minded practices with hands-on coding, making complex statistical outputs tidy and reproducible for downstream users. He also brings a curious, principled approach to testing and dataset choices, demonstrated by thoughtful updates that keep examples and edge-case handling aligned with real-world workflows.
8 years of coding experience
Johns Hopkins University
Stackoverflow
501reputation
4kreached
20answers
0questions
Github Skills (17)
r10
testing10
package-development10
data-preprocessing10
statistical-models10
data-analysis10
feature-engineering9
documentation9
data-science8
cran6
tidymodels6
h2o6
lightgbm6
machine-learning6
hyperparameters6
Programming languages (13)
Github contributions (5)
tidymodels/broom
May 2020 - Jan 2023
Convert statistical analysis objects from R into tidy format
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:21 releases, 26 reviews, 635 commits in 2 years 8 months
Contributions summary:Simon's contributions primarily revolve around improving the `broom` package, which focuses on converting statistical analysis objects from R into a tidy format. The commits demonstrate the implementation of functionality to handle and document new arguments for existing tidiers, indicating an understanding of the package's internals. Furthermore, the user worked on improving the handling of models from the `MASS` and `rstanarm` packages, as well as the `glance` method for objects that are part of complex survey data, indicating the ability to work in the area of data science and model integration in R. The user also contributes to the refactoring of existing code to use more modern and more performant libraries.
r-packagestatisticshypothesis-testingregression-modelstidymodels
tidymodels/recipes
Apr 2022 - May 2022
Pipeable steps for feature engineering and data preprocessing to prepare for modeling
Role in this project:Data Scientist
Data ScientistContributions:129 reviews, 8 commits, 14 PRs in 7 days
Contributions summary:Simon primarily focused on modifying and testing the `recipes` package, specifically related to feature engineering and data preprocessing. Their work involved transitioning example datasets from `okc` to `Sacramento`, indicating a focus on adapting and validating existing functionality. The user made adjustments to test cases, including dummy variables and NA handling, demonstrating a commitment to ensuring data preparation steps function correctly.
pythondata-preprocessingmachine-learningengineeringprepare
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.