Maja Ritarudolph
Research Assistant at Columbia University
United States
Summary
🤩
Rockstar🎓
Top SchoolMaja Ritarudolph is a research-focused software engineer and PhD candidate in Electrical Engineering at Columbia University with 14 years of experience in probabilistic modeling, variational inference, and recommender systems. As a Research Assistant under Prof. Shih-Fu Chang she combines rigorous mathematical training (B.S. in Mathematics from MIT) with hands-on implementation of inference methods and applied ML systems. Her open-source contributions include implementing and testing MAP estimation in the widely used Edward probabilistic programming library, demonstrating practical expertise in TensorFlow-based deep generative models. Early internships and research roles span bioinformatics, EEG-based brain-computer interfaces, and theoretical mathematics, reflecting a rare blend of applied and theoretical strengths. She has a track record of turning statistical theory into reproducible code and educational materials, and she brings attention to numerical validation and example-driven testing in research software. Colleagues can expect a meticulous researcher who bridges clean mathematical thinking with production-aware software development.
14 years of coding experience
1 year of employment as a software developer
B.S., Mathematics, 4.7 / 5.0, B.S., Mathematics, 4.7 / 5.0 at Massachusetts Institute of TechnologyPhD, Electrical Engineering, PhD, Electrical Engineering at Columbia University in the City of New YorkMathematics, 4.0 / 4.0, Mathematics, 4.0 / 4.0 at Delaware State University
Github Skills (9)
statistics10
bayesian-methods10
machine-learning10
variational-inference10
probabilistic-programming10
tensorflow10
python10
deep-learning9
neural-network8
Programming languages (3)
Github contributions (5)
blei-lab/edward
Mar 2016 - Mar 2017
A probabilistic programming language in TensorFlow. Deep generative models, variational inference.
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:68 commits, 13 PRs, 65 pushes in 1 year
Contributions summary:Maja implemented and tested Maximum a Posteriori (MAP) estimation within the `edward` probabilistic programming library. The contributions involved modifying existing inference methods (MFVI) and variational families (MFPointMass) to support MAP estimation. Additionally, the user fixed issues related to MAP and other variational methods, as well as created examples for Gaussian and Beta-Bernoulli models. The user also adapted existing examples in `examples/` to utilize the `MAP` class and test its functionality.
inferenceneural-networksmachine-learningprobabilistic-programmingdeep-generative-models
TalkingData/edward
Jun 2016 - Jul 2016
Contributions:24 commits in 1 month
deep-learningprobabilistic-modelinginferencemachine-learningruns
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.