Sergey Orlov
Data Scientist @Cashco Financial
Signup · Get unlimited contacts
WORK HISTORY
Data Scientist @Cashco Financial
Calgary, AB, CA
Designed, trained, and deployed a default prediction model using XGBoost on transactional and credit bureau data (Inverite, Flinks, TransUnion), improving recall by 35% while maintaining high precision. The model reduced the default rate on high-risk clients from 16% to 6%, directly impacting portfolio profitability.• Engineered domain-specific behavioral features (e.g, gambling activity, overdraft frequency, payday loan usage) from large-scale transaction records, significantly boosting predictive accuracy and model interpretability.• Built a production-ready Python pipeline for automated model scoring, SHAP-based explainability, and real-time/batch risk evaluation, enabling underwriters and portfolio managers to make data-driven lending decisions.• Designed and implemented ETL pipelines and PySpark notebooks in Microsoft Fabric, consolidating data from four sources (MS SQL, MySQL, PostgreSQL) into a single, trusted layer for analytics and modelling.• Developed monitoring and decision-review dashboards in Tableau and Power BI, providing executives with actionable insights on portfolio risk exposure.• Processed datasets with 40K+ loans and 130+ features, applying time-based aggregations, JSON parsing, and financial pattern detection to enhance data quality and feature richness.• Contributed to the architecture of a hybrid decision engine (rules + ML), ensuring model decisions align with regulatory requirements and business constraints in high-stakes lending workflow.
EDUCATION
University of Calgary
Master of Science - MS, Data Science and Analytics
Tomsk State University
Bachelor's degree, Mechanical Engineering
Tomsk State University
Bachelor's degree, Management of Organization
Tomsk State University
Doctor of Philosophy - PhD, Physics and Mathematics
Tomsk State University
Master of Engineering - MEng, Applied Mechanics
ABOUT SERGEY ORLOV
I’m a Data Scientist & Machine Learning Engineer with a Ph.D. in Physics and Mathematics and 8+ years of experience delivering data-driven solutions across finance, technology, and engineering domains. With 15 years in High-Performance Computing and a decade leading cross-functional tech teams, I bridge the gap between research, engineering, and business — turning complex data into real business value. I specialize in building and deploying scalable machine learning models, designing efficient data architectures, and leading end-to-end analytics initiatives that drive measurable impact — from data ingestion and modelling to actionable insights and automation. Core strengths: • Machine Learning, Predictive Modelling, and LLMs • Data Engineering (ETL, Spark, Dask, SQL/NoSQL, Data Pipelines) • Statistical Analysis, Feature Engineering, and Experimentation • Cloud & Infrastructure: Azure, AWS, GCP, Docker, Git • Visualization & Communication: Power BI, Tableau, Plotly, Streamlit • Leadership in Data Teams, Project Management, and Technical Strategy Passionate about using data to make smarter business decisions and empowering teams to deliver measurable outcomes — not just models. Always open to discussing opportunities where data science meets business growth.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.