Machine Learning











Senior Data Scientist


Dec 2021 – Present Berlin
  • Competence Lead of the team of data scientists who drive efforts for fraud prevention in Klarna core products in key EU and AP markets.
  • Core responsibilities: end-to-end model building, stakeholder management, quarterly planning, cross-domain communication, professional development of the team.
  • Fraud model creation via supervised and unsupervised frameworks.

Data Scientist


Sep 2020 – Nov 2022 Berlin
  • Developing real-time bidding framework based on conversion rate.
  • Data-driven attribution modeling based on Shapley value framework.
  • Working with GC AI Platform, building BigQuery ETL for predicting customer behavior.

Data Scientist


Sep 2019 – Sep 2020 Berlin
  • Implementing real-time ML fraud detection framework: reducing the false positive rate in fraud predictions to maximize profit, designing KPIs for evaluating fraud predictive models.
  • Designing and implementing return prediction ML framework for e-commerce.

Research Assistant / Data Scientist

Telekom Innovation Laboratories: T-Labs

Mar 2017 – Sep 2019 Berlin
  • Worked on the German-Hungarian project in digital finance eBiz:

    • data analysis of financial transactional real-world networks
    • developing a recommendation solution for small and medium enterprises, allowing partner discovery using ML APRIORI frequent itemset mining and business relationship graphs
    • design and implementation of prototypes using RShiny
    • pattern mining on unlabelled transactional real-world data
  • Fraud and anomalies detection:

    • detection of fraudulent Ponzi and Pyramid schemes in financial networks
    • unsupervised node segmentation in financial transactional networks
    • solvency analysis of financial network participants using ARIMA time series forecasting

Project Manager

various companies

Sep 2012 – Oct 2016
  • Managed several engineering teams, launched over 10 projects of social public e-services for citizens, mobile applications, integrations with third-party information systems serving for the audience of over 200k users.
  • More details about my career path can be found in my resume.



Annual Salary Survey Analysis

Analysis & dynamic visualizations of the public anonymous salary survey data.

Multithreading Scraping

Web scraping using python multithreading and a real-time-updated list of available proxy servers.

Twitter users geospatial analysis

Visualisation of the Twitter users on the map.

Network Embeddings

Customization of Node2vec algorithm with time awareness for Small-world networks.

Recommender System

Recommender System to recommend loyalty cards to customers.

Podcasts & Articles & Open Source

Long term analysis & dynamic visualizations of salary trends in Germany for 7 years.

Regularization in Regression - remedy for numerical instability.

scikit-uplift is an uplift modeling python package that provides fast sklearn-style models implementation, evaluation metrics and visualization tools.

Career tips and tricks

Contact me