Ksenia Legostay
Data Science Manager
Berlin, Germany
Summary
👤
Senior🎓
Top SchoolKsenia Legostay is a Data Science Manager at Klarna with 9 years of FinTech experience specializing in fraud prevention and risk-focused ML. She builds production-grade models and leads cross-functional initiatives that measurably improve risk metrics, while also standardizing best practices across feature engineering, MLOps, deployment, monitoring, and interpretability. As a competence lead and mentor she drives technical excellence through code reviews, mentoring, and community engagement, and she regularly speaks at industry events. Ksenia blends hands-on contributions—such as improving uplift modeling tooling and tests in an open-source scikit-uplift project—with strategic stakeholder alignment across product, compliance, and operations. Based in Berlin, she pairs an engineering background and a Master’s in Management Information Systems with a pragmatic focus on turning complex model insights into clear, actionable strategies. Her work is notable for closing the loop between rigorous research-grade methods and robust, auditable production outcomes.
9 years of coding experience
7 years of employment as a software developer
Master’s Degree Management Information Systems General, Master’s Degree Management Information Systems General at Technische Universität BerlinExpert Business Administration and Management General, Expert Business Administration and Management General at Kazan State University
English, German, Russian
Github Skills (12)
lift10
scikit-learn10
scikit10
machine-learning10
lifting10
python10
modeling10
data-analysis10
causal-inference9
testing9
pandas8
numpy8
Programming languages (4)
Github contributions (5)
maks-sh/scikit-uplift
Jul 2021 - Jul 2021
:exclamation: uplift modeling in scikit-learn style in python :snake:
Role in this project:
Data Scientist
Data ScientistContributions:5 commits, 7 PRs, 17 comments in 4 days
Contributions summary:Ksenia made several contributions focused on enhancing the `scikit-uplift` library, which is centered on uplift modeling. The user added binary target checkers to metrics functions and visualization code, contributing to the library's robustness and correctness. Moreover, the user added tests for the `fetch_lenta` and `fetch_x5` datasets, which are essential for verifying the library's functionality and data handling capabilities. The user's work involved direct interaction with the core components of the library.
net-liftpythondata-sciencesnakeuplift-modeling
Ksyula/Salary-report
Jan 2020 - Jan 2023
Salary report analysis & visualization with pandas and plotly
Contributions:12 commits, 10 PRs, 65 pushes in 3 years 1 month
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.