Adam Kelleher
Adjunct Associate Professor
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolAdam Kelleher is an applied data scientist and academic with 13 years of experience bridging causal inference research and production analytics, currently teaching at Columbia's Data Science Institute while leading alternative-data research as Chief Data Scientist at Barclays. He focuses on recommender systems, virality, and information diffusion, and brings a strong math and physics background (PhD in theoretical gravity/cosmology) to large-scale social data problems. Adam is an active open-source contributor—helping build DoWhy features and maintaining causality tooling (pip install causality)—that surfaces Pearlian causal models and classical estimators for practitioners. He has a track record of shipping ML systems in industry (BuzzFeed) and developing curriculum to train students in real-world causal methods, combining rigorous theory with pragmatic tooling. An interesting wrinkle: he moved from building distributed web crawlers and physics labs to creating fast graph visualizations (pyh3) and a Python causal-analysis ecosystem, showing a rare blend of engineering, teaching, and foundational research.
13 years of coding experience
11 years of employment as a software developer
BS, Physics (mathematics), BS, Physics (mathematics) at Clemson UniversityDoctor of Philosophy (Ph.D.), Theoretical Gravity/Cosmology, Doctor of Philosophy (Ph.D.), Theoretical Gravity/Cosmology at The University of North Carolina at Chapel HillMS, Physics (Cosmology), MS, Physics (Cosmology) at University of North Carolina at Chapel Hill
Github Skills (14)
causal10
pandas10
machine-learning10
causality10
python10
data-science10
causal-inference10
apidoc9
statistical-models9
scipy9
api9
statistics9
bootstrapping8
unit-testing8
Programming languages (3)
Github contributions (5)
akelleh/causality
May 2015 - Nov 2021
Tools for causal analysis
Role in this project:
Data Scientist
Data ScientistContributions:4 releases, 191 commits, 104 PRs in 6 years 6 months
Contributions summary:Adam implemented and tested several functions for causal analysis, including discrete and continuous expectation calculations, mutual information estimation, and the calculation of p-values. They added a `DataSet` object and a `MutualInformation` class to facilitate these analyses. The user also added test coverage for the various features.
causalcausal-inferencegraphscausal-analysis
py-why/dowhy
Jan 2019 - May 2019
DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions. DoWhy is based on a unified language for causal inference, combining causal graphical models and potential outcomes frameworks.
Role in this project:Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:35 commits, 13 PRs, 25 comments in 4 months
Contributions summary:Adam primarily contributed to the development of the DoWhy Python library, focusing on causal inference. They implemented a DataFrame API for causal analysis, enabling the use of pandas DataFrames within the library. Additionally, the user added support for estimator runs, including the G-Formula estimator, and built the base for other estimators. Finally, the user added test coverage to the causal estimators.
causal-inferencedowhygraphical-modelspythontesting
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.