Here is a resume pdf with lots of impact numbers and other juicy details. I also still have a CV.
Amazon 2023+
I work on fraud detection and mitigation for Kindle Unlimited. Mostly I've worked on real-time royalty fraud models and infrastructure, as well as fraud tagging systems and general ML infrastructure (data/model orchestration).
Meta 2021-2023
I was a central researcher (CAS) embedded in Business Integrity (ads) ensuring cross-team problems like appeal abuse got solved and automatic security friction (like captcha, 2fac) was triggered intelligently. I also spent a lot of time trying to get people to do experiments rather than assuming "obvious" things.
Institute for Health Metrics and Evaluation 2019-2021
I was a consultant, mostly producing custom forecasts using a hierarchical forecasting system I developed which mostly ripped off HTS but in Python. It did things like produce hierarchically consistent forecasts with uncertainty estimates.
Postdoctoral Fellow eScience Institute 2017-2019
I continued doing ML interpretability research and software development.
Google Summer of Code Fellow 2015
I worked on a model agnostic feature importance and model interpretation module for MLR with the eminent Bernd Bischl and helped integrate it with BayesOpt: also a bit on using ML models for BayesOpt instead of GP. It was genuinely a great time.
Pennsylvania State University 2014-2017
I worked on network modeling and causal inference research with my advisor Bruce as well as more work on interpretability software, and procedures to improve reproducibility.
University of Georgia 2011-2014
I applied ML to quantitative political violence research (the APSR) and also analyzed the patterns of marijuana prices for insight into trafficking patterns.