Larry Dong
Data Scientist
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolLarry Dong is a data scientist with nine years of experience blending Bayesian statistics, probabilistic programming, and applied biostatistics to solve real-world problems. He contributed core functionality to PyMC and Aesara—adding new distributions and performance-focused refactors—and helped grow pymc-marketing into a widely used Bayesian marketing toolbox by implementing CLV models and visualization utilities. His research background spans clinical trial reanalysis and treatment optimization, and he recently explored using LLMs for latent variable inference as a JSPS Research Fellow at The University of Tokyo. Currently at Roche and with a PhD in Biostatistics from the University of Toronto, he moves comfortably between research, teaching, and production-grade open-source development. Notably, his open-source work shows both low-level backend improvements (shape inference, ops refactors) and high-level model building, reflecting a rare mix of numerical rigor and practical product impact.
9 years of coding experience
3 years of employment as a software developer
Diploma of College Studies Honors Health Sciences, Diploma of College Studies Honors Health Sciences at Marianopolis CollegeDoctor of Philosophy - PhD Biostatistics, Doctor of Philosophy - PhD Biostatistics at University of TorontoPublic Health Data Science Public Health, Public Health Data Science Public Health at Institut de santé publique, d'épidémiologie et de développement (ISPED)Master's degree Biostatistics, Master's degree Biostatistics at McGill University
English, French, Chinese
Stackoverflow
1reputation
693reached
0answers
2questions
Github Skills (24)
probabilistic-programming10
python10
optimizers10
data-science10
testing10
statistics10
optimizer10
statistic10
numpy10
mcmc10
automatic-differentiation10
pymc10
bayesian10
compiler10
bayesian-inference10
Programming languages (7)
Github contributions (5)
pymc-labs/pymc-marketing
Jun 2022 - Mar 2023
Bayesian marketing toolbox in PyMC. Media Mix (MMM), customer lifetime value (CLV), buy-till-you-die (BTYD) models and more.
Role in this project:
Data Scientist
Data ScientistContributions:42 reviews, 10 commits, 8 PRs in 8 months
Contributions summary:Larry contributed significantly to the development of Bayesian marketing models within the repository. Their work involved implementing distribution classes for both continuous non-contractual and contractual settings, demonstrating an understanding of statistical modeling. They also added and tested a BetaGeoModel to the clv submodule, including associated tests and notebooks, suggesting a focus on customer lifetime value analysis. Furthermore, they integrated plotting functionalities for CLV models, indicating an effort to visualize and interpret the results.
pythonclvpymctoolboxmarketing
pymc-devs/pymc
Apr 2021 - Oct 2022
Bayesian Modeling and Probabilistic Programming in Python
Role in this project:Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:195 reviews, 21 commits, 40 PRs in 1 year 6 months
Contributions summary:Larry primarily worked on the PyMC library, addressing deprecation warnings related to NumPy data types and fixing grammar. They also contributed to the addition of new distributions and moments, specifically focusing on the Dirichlet, Interpolated, and StickBreakingWeights distributions. Furthermore, the user modified existing code related to moment calculations, which are essential in the Bayesian modeling context, and incorporated new test cases. This work aligns with the repository's focus on probabilistic programming and Bayesian inference.
pythonbayesian-inferencestatistical-inferencemachine-learningprobabilistic-programming
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.