Projects
Open source software
- ARTKIT - Python library for Automated Red Teaming (ART) of generative AI applications
- I am a core developer and maintainer of ARTKIT, which was developed at BCG X as part of our Responsible AI program.
- Read more about the development of ARTKIT on the BCG X Blog.
- Check out the Official Docs for a comprehensive user guide and tutorials.
- btw - R wrapper for BayesTraitsV3
- BayesTraits is a popular academic software for Bayesian phylogenetic comparative methods developed by Mark Pagel and Andrew Meade at the University of Reading)
btw
facilitates automation of BayesTraits workflows with R scripting and has been used in dozens academic projects
- caRds - The most important work I’ve ever done
Independent analytics projects
Olympic history
- Kaggle dataset - I scraped and cleaned a complete historical dataset of Olympic athletes and medal results from www.sports-reference.com. git
- This dataset has been downloaded from Kaggle over 195K times as of January 2025.
- Kaggle kernel - I published an extensive analysis to accompany the dataset, focusing on exploratory analysis and visualization with R.
- This analysis has been viewed +100k times and copied/downloaded +800 times as of January 2025.
Twitter analytics
- Twitterstorm dynamics - I pulled ~5K tweets from a Twitterstorm that embroiled a colleague and did a deep dive into the dynamics of the storm. git
- NHL vs female Olympian salaries - I built a model of
Salary ~ Twitter popularity
for NHL hockey players and showed that if female U.S. Olympic hockey players were subjected to the same model, they would be millionaires. git