Contact information

You can get this candidate's contact information from https://www.work.ua/resumes/19953592/

Work experience

Data Scientist

from 10.2025 to now (1 year)
King Group, Київ (IT)

Data Scientist / Machine Learning Engineer — Production ML & MLOps
• Designed and productionized ML solutions for player-behavior forecasting, deposit and withdrawal prediction, churn prevention,
personalized game recommendations, customer segmentation, and anomaly detection in business metrics.
• Translated business questions into ML designs by defining populations, prediction points, as-of semantics, feature/label
windows, forecast horizons, temporal splits, and measurable success criteria.
• Built BigQuery feature pipelines over behavioral, gaming, payment, bonus, withdrawal, profile, and marketing data; implemented
validation for schemas, duplicate keys, join cardinality, label maturity, and output completeness.
• Developed classification, regression, ranking, time-series, and clustering models using XGBoost, LightGBM, scikit-learn, Optuna,
LambdaRank, Prophet, and collaborative/content-based recommendation methods.
• Converted notebook research into configurable Kedro pipelines with shared train/validation/inference transformations and
independent models, configurations, and artifacts for multiple projects and countries.
• Deployed daily batch scoring and retraining on Google Cloud using Vertex AI Custom Jobs, Model Registry, Vertex Experiments,
Cloud Run Jobs, Cloud Scheduler, GCS, Artifact Registry, and Docker.
• Engineered reliable BigQuery publishing with staging tables, idempotent MERGE/upsert, latest/history datasets, two-phase writes
for parallel jobs, versioned artifacts, TTL retraining policies, checkpoints, and safe historical backfills.
• Performed rolling and time-based evaluation using ROC-AUC, PR-AUC, calibration, Brier score, F1, ranking metrics, lift,
coverage, and stability by time and segment; audited pipelines for leakage and train/production contract mismatches.
• Documented data contracts, runbooks, architecture, and model results for analytics, retention, marketing, and management
stakeholders.

Data Scientist

from 11.2024 to 10.2025 (1 year)
OTP Bank, Київ (IT)

Data Scientist / Machine Learning Engineer
• Developed and maintained classification, regression, clustering, and scoring models for internal banking products and
department-specific business needs.
• Scraped and parsed external data, prepared datasets for movable and immovable property valuation, and built predictive models
on the resulting data.
• Performed customer analytics and cluster profiling, improved internal datasets, automated recurring scoring workflows, and
communicated performance through reports and visualizations.
• Worked with large-scale data and collaborated with business stakeholders to translate analytical requests into reusable
modeling solutions.

Data scientist

from 12.2022 to 12.2024 (2 years)
Cітон Груп, ТОВ, Київ (IT)

Computer Systems Analyst
• Built inventory and preprocessing components for monitoring data and collaborated with the wider development team.
• Developed ML-based monitoring solutions for root-cause analysis, anomaly detection, adaptive thresholds, and forecasting.
• Automated SAS marketing workflows and daily campaign reporting.

Education

Київський політехнічний інститут імені Ігоря Сікорського

ІПСА, Системи та методи штучного інтелекту, Київ
Higher, from 2023 to 2025 (2 years)

Київський політехнічний інститут імені Ігоря Сікорського

ІПСА, Системний аналіз, Київ
Unfinished higher, from 2019 to 2023 (4 years)

Additional education and certificates

Системна математика

2023-2025

Knowledge and skills

• Python
• Google Cloud Platform
• PyTorch
• Scikit-learn
• TensorFlow
• SciPy
• NumPy
• Oracle SQL Developer
• BQ
• Pandas
• MS Excel
• Deep Learning
• Machine learning
• C++
• Keras
• Kedro
• Vertex
• ML Ops

Language proficiencies

• English — above average
• Ukrainian — fluent

Additional information

Commercial work experience (3 years 8 months):

Machine Learning Engineer / Data Scientist

Currently working as a Machine Learning Engineer / Data Scientist, developing end-to-end production ML systems for player-behavior forecasting, churn prediction, deposit and withdrawal prediction, personalized recommendations, customer segmentation, and anomaly detection.

Responsible for the complete ML lifecycle: translating business requirements into ML tasks, defining temporal feature and label windows, building BigQuery feature pipelines, training and evaluating models, converting research notebooks into configurable Kedro pipelines, and deploying automated scoring, retraining, monitoring, and backfill processes.

Developed classification, regression, ranking, clustering, recommendation, and time-series solutions using XGBoost, LightGBM, scikit-learn, Optuna, LambdaRank, ItemKNN, implicit ALS, Prophet, and deep learning methods. Built multi-project and multi-country pipelines with independent configurations, models, and versioned artifacts.

Deployed ML workloads using Vertex AI, Cloud Run Jobs, Cloud Scheduler, BigQuery, GCS, Artifact Registry, Docker, and Kedro. Implemented schema validation, staging tables, idempotent MERGE operations, latest/history tables, model versioning, and safe historical backfills.

Performed temporal validation, rolling backtesting, calibration, and model evaluation using ROC-AUC, PR-AUC, F1, Brier score, lift, coverage, and ranking metrics. Conducted ML and data-quality audits covering data leakage, label maturity, temporal windows, duplicate keys, incorrect joins, and train/inference consistency.

Previously worked at OTP Bank, developing classification, regression, clustering, and scoring models for banking products. Focus areas included property valuation, customer analytics, segmentation, and predictive modeling of customer behavior. Automated daily and monthly scoring workflows, prepared datasets, and delivered model-performance reports and visualizations to stakeholders.

Earlier experience at Seeton included building monitoring-data systems, preprocessing components, and ML-based solutions for root-cause analysis, anomaly detection, adaptive thresholds, and forecasting. Also automated SAS marketing workflows and daily campaign reporting.