Code Repository for "PCS Workflow for Veridical Data Science in the Age of AI"

This repository contains code and data corresponding to:

Zachary T. Rewolinski, Bin Yu. "PCS Workflow for Veridical Data Science in the Age of AI." Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences. 2025.

Specifically, the code and data included in this repository include everything needed to reproduce our analysis and figures from the case study in Section 9.

Directory Structure

• code/case-study.ipynb: Jupyter Notebook containing the case study example from Section 9.
• deidentified-cleaned-data/: anonymized cleaned datasets from the 19 students who gave us explicit permission.
• figures/: figures from the manuscript.
• original-data/: the raw, uncleaned PECARN ciTBI dataset and the corresponding data documentation.
• workflow.yml: conda environment used by the authors, included for version control and reproducibility.

Citation

While we wait for the citation of the published article, please use the following preprint citation:

@misc{rewolinski2025pcsworkflow,
      title={PCS Workflow for Veridical Data Science in the Age of AI}, 
      author={Zachary T. Rewolinski and Bin Yu},
      year={2025},
      eprint={2508.00835},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2508.00835}, 
}