Jeffrey Arnold
Senior Data Scientist, Causal Inference
Oakland, California, United States
Summary
🤩
Rockstar🎓
Top SchoolJeffrey Arnold is a senior data scientist specializing in causal inference and experimentation with 16 years of experience building production-ready ML and Bayesian systems at companies like Discord and Instacart. He blends an academic background (PhD) and faculty experience with hands-on engineering—designing experiment platforms, automated bidding systems, and scalable estimators implemented in Python, Go, Scala, Snowflake and Stan. Jeffrey is an avid open-source contributor to the R ecosystem and Stan (notably improving rstan, ggplot2, readr, knitr and tidymodels), reflecting deep expertise in statistical modeling, reproducible workflows, and time-series/Bayesian methods. He’s shipped features that measurably moved business metrics (e.g., ad auto-bidding and deals ranking) and holds a patent-level hashing design for experimentation infrastructure. Based in Oakland, he also ships full-stack prototypes and AI tooling, having bootstrapped technical products at early-stage startups and research labs.
17 years of coding experience
10 years of employment as a software developer
BA Economics Government, BA Economics Government at Dartmouth CollegePhD Political Science, PhD Political Science at University of Rochester
Stackoverflow
126reputation
5kreached
3answers
0questions
Github Skills (42)
data-visualizations10
bayesian-statistics10
javascript10
restructuredtext10
color-theory10
r10
data-visualisation10
feature-engineering10
data-preprocessing10
knitr10
ace-editor10
dplyr10
bayesian10
statistical-models10
parsing10
Programming languages (13)
Github contributions (5)
jrnold/ggthemes
Sep 2012 - Jan 2021
Additional themes, scales, and geoms for ggplot2
Role in this project:
Data Scientist
Data ScientistContributions:1 review, 634 commits, 75 PRs in 8 years 5 months
Contributions summary:Jeffrey's commits focus on adding new shape scales and implementing mapping from Stata color names to RGB hex values. These changes involve incorporating visual elements and color schemes related to data visualization techniques, specifically from the work of Cleveland. The user also contributed to expanding the functionality of the library by including tools for solarized colors and additional color palettes for use with various libraries such as Stata and LibreOffice.
ggplotplotggplot2visualizationdata-visualisation
jrnold/r4ds-exercise-solutions
Dec 2016 - Jul 2020
Exercise solutions to "R for Data Science"
Role in this project:Data Scientist
Data ScientistContributions:978 commits, 129 PRs, 952 pushes in 3 years 7 months
Contributions summary:Jeffrey worked on a model in the "model" section by doing "a little more work." They focused on the model section by including file ch23.nb.html, as well as updating and creating ch20.nb.html and the related figures. These contributions demonstrate a focus on model development and refinement, likely indicating tasks related to data analysis, feature engineering, or model training.
data-scienceexercise-solutionstidyverserggplot2
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.