Irina Elisova
Head Of Machine Learning at GEOMOTIVE
Uzbekistan
Summary
👤
Senior🎓
Top SchoolIrina Elisova is a Head of Machine Learning with over 10 years of experience building production AI and recommender systems, currently leading ML at GEOMOTIVE and lecturing on RecSys at ITMO University. She has driven measurable product impact—boosting engagement and CTR through uplift, look-alike, and hybrid cold/hot recommendation models—and built A/B testing and experimentation practices used across teams. A prolific open-source contributor and co-author of two RecSys courses, she is the top maintainer of scikit-uplift, having added visualization and weighted-uplift metrics to a library with hundreds of stars and real production adopters. Her background spans hands-on PySpark/Python engineering, platform design, and teaching 300+ master’s students, blending rigorous applied mathematics training with practical product delivery. Notably, she translated research-grade uplift methodology into accessible tooling and tutorials that reached tens of thousands of readers and practitioners.
10 years of coding experience
7 years of employment as a software developer
Master's degree Applied Mathematics, Master's degree Applied Mathematics at Moscow Aviation Institute (National Research University)
Russian, English
Github Skills (13)
scikit10
data-visualizations10
lift10
data-visualization10
data-visualisation10
lifting10
python10
modeling10
scikit-learn10
matplotlib9
machine-learning9
pandas8
causal-inference8
Programming languages (2)
Github contributions (5)
maks-sh/scikit-uplift
Apr 2020 - Jun 2022
:exclamation: uplift modeling in scikit-learn style in python :snake:
Role in this project:
Data Scientist
Data ScientistContributions:19 reviews, 11 commits, 22 PRs in 2 years 2 months
Contributions summary:Irina primarily contributed to the implementation of uplift modeling visualization tools within the `scikit-uplift` library. They added and refined the `plot_uplift_by_percentile` function, expanding its capabilities with different plot types (line and bar) and improved its presentation. Furthermore, they integrated the `weighted_average_uplift` metric and other related features to enhance the analysis of uplift model performance. Their work focused on extending the library's functionality for evaluating and visualizing uplift results.
net-liftpythondata-sciencesnakeuplift-modeling
ElisovaIra/scikit-uplift
Apr 2020 - Mar 2021
:exclamation: uplift modeling in scikit-learn style in python :snake:
Contributions:37 PRs, 34 pushes, 10 branches in 10 months
pythondata-sciencesnakeuplift-modelingmachine-learning
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.