Colin Carroll
Technical Staff at Anthropic
Somerville, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolColin Carroll is a technical staff engineer with 12 years of experience building probabilistic modeling and ML infrastructure, currently based in Somerville, MA and working at Anthropic after roles at Google and DeepMind. He combines a Ph.D. in Mathematics with hands-on software and DevOps skills, shipping improvements to core Bayesian tooling like PyMC, TensorFlow Probability (including a Dual Averaging HMC step-size implementation), and ArviZ. His background spans data science, backend engineering, and test/build infrastructure—reflecting an unusual blend of rigorous mathematical training and production-grade engineering. Colin is an active open-source contributor who has touched both algorithmic cores (MCMC samplers, distributions) and developer experience (install/test scripts, linting, docs). He has a track record of improving model reliability and developer workflows at scale, informed by earlier analytics roles where he moved large datasets into actionable insights. Expect someone who bridges deep theory and pragmatic engineering to make probabilistic methods more robust and usable in production.
12 years of coding experience
13 years of employment as a software developer
Ph.D. Mathematics, Ph.D. Mathematics at Rice UniversityBachelor of Arts Mathematics and Economics, Bachelor of Arts Mathematics and Economics at Williams College
Github Skills (32)
data-visualizations10
probabilistic-programming10
mc10
python10
matplotlib10
testing10
statistics10
codelist10
machine-learning10
data-visualisation10
exploratory-data-analysis10
tensorflow-probability10
bayesian10
tensorflow10
lint10
Programming languages (12)
Github contributions (5)
pymc-devs/pymc4
Jun 2018 - Dec 2019
Experimental PyMC interface for TensorFlow Probability. Official work on this project has been discontinued.
Role in this project:
Data Scientist / ML Engineer
Data Scientist / ML EngineerContributions:31 commits, 42 PRs, 34 pushes in 1 year 7 months
Contributions summary:Colin contributed to the development of a PyMC4 interface for TensorFlow Probability, which suggests a focus on probabilistic programming and statistical modeling. Their commits include storing and using model arguments, modifying the forward sample return type, adding key methods like `__getitem__` to random variables, and refactoring contexts for model inference. Furthermore, the user added functionality for Multinomial and Dirichlet distributions, which are fundamental components of many statistical models. This user appears to be working on core functionalities of this project.
pythonpymcmachine-learningdiscontinuedprobability
arviz-devs/arviz
May 2018 - Sep 2022
Exploratory analysis of Bayesian models with Python
Role in this project:Back-end Developer & DevOps Engineer
Back-end Developer & DevOps EngineerContributions:2 releases, 17 reviews, 181 commits in 4 years 5 months
Contributions summary:Colin's commits primarily focus on improving the codebase's structure and maintainability. They added a linting framework and made modifications to existing code. The user also implemented changes in arviz/stats/stats.py, suggesting interaction with the core statistical functions of the library, and also focused on refactoring the plotting modules of the library. Their work suggests an effort towards standardizing and improving the quality and consistency of the codebase.
pythonexploratorybayesian-inferencemachine-learningbayesian-models
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.