Description

I am a Data Scientist and ML Engineer with a strong focus on geo analytics and business-driven machine learning solutions.

I help companies turn location data into actionable insights — from demand forecasting and site selection to customer behavior analysis and risk modeling. I combine machine learning with geospatial data (OSM, mobility, demographics) to solve real-world business problems, especially in retail, urban analytics, and fintech.

In my work, I build end-to-end solutions: from data collection and feature engineering to modeling, validation, and deployment. I have experience with Python, SQL, PyTorch, and modern ML pipelines, including boosting models and computer vision.

I am also the founder of GeoSurf, a geoanalytics platform focused on predicting consumer activity across locations and supporting data-driven decision-making for businesses and public sector.

I prefer working on projects where machine learning directly impacts business metrics — revenue, risk, or operational efficiency.

Industry field of expertise

Languages

Russian
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Batumi (up to 50km)

Yandex Fintech
Senior Data Scientist
BANKING AND INSURANCE
January 2026 - Today (5 months)
Moscow, Russia
Contributed to the development of B2B credit products and credit risk optimization, with a focus on data enrichment, credit limit strategy, and portfolio growth.
- Integrated new external and internal data sources into the B2B credit decisioning process, expanding the data coverage for borrower assessment and improving the quality of credit risk evaluation
- Improved credit decisioning by enhancing client segmentation and risk assessment logic, which helped increase approval accuracy while keeping risk within target levels
- Supported the credit limit increase initiative for selected B2B client segments, enabling higher available limits for reliable borrowers based on new sources
- Collaborated with product, risk, analytics, and data engineering teams to launch new data pipelines and operationalize updated credit strategies
- Participated in the refinement of scoring and limit assignment approaches, allowing the business to better identify low-risk, high-potential clients
- Built analytics for new data sources to support credit limit assessment and improve B2B credit decisioning
Data science SQL Python Credit scoring CI/CD
GeoSurf,
CEO & Founder
July 2024 - Today (1 year and 11 months)
Moscow, Russia
As the I lead the development of an AI-powered geospatial analytics platform, turning big data into actionable insights for businesses
Sovcombank
Data Scientist
BANKING AND INSURANCE
May 2023 - June 2025 (2 years and 1 month)
Developed ML models for antifraud, credit scoring, bankruptcy prediction, took part in FL and Videoanalytics projects. Owned the full model development lifecycle from data collection and feature engineering to validation and production deployment.

- Developed session-level and transaction-level antifraud models to detect suspicious user behavior and reduce fraud-related risks
- Developed a bankruptcy prediction model to identify high-risk clients and support proactive risk management
- The bankruptcy prediction model generated a multi-million / multi-billion RUB business impact while helping reduce potential credit losses
- Developed revenue prediction models to estimate expected model-driven income and support business decision-making
- Participated in a video analytics project, applying ML / deep learning approaches to visual data processing and analysis
- Collected and prepared large-scale datasets using PySpark and SQL
- Designed and implemented feature engineering pipelines using Python and pandas
- Performed feature selection, model training, validation, and performance analysis
- Built and compared models using CatBoost, XGBoost, LightGBM, scikit-learn, and PyTorch
- Evaluated model quality on test and out-of-time samples to ensure stability and generalization
- Supported production deployment of ML models and collaborated with engineering teams on implementation

Tech stack: Python, SQL, PySpark, YOLO, pandas, NumPy, scikit-learn, CatBoost, XGBoost, LightGBM, PyTorch, feature engineering, feature selection, model validation, OOT validation, antifraud, credit scoring, bankruptcy prediction, revenue prediction, video analytics