Mykola Symotiuk

Created, cleaned, and labeled large-scale video datasets for computer vision tasks. Fine-tuned and debugged pre-trained object-detection models; developed backend Python applications (Flask, FastAPI) with scalable deployment on Azure Cloud and Cloud Foundry. Built and optimized graph database pipelines (Neo4j) for ML-driven analytics; engineered automated data transfers into graph databases. Collaborated with cross-functional teams to integrate ML-powered solutions into operational pipelines, improving efficiency and scalability.

Collected, cleaned, and structured large-scale transactional and cash-point datasets for analysis. Conducted time-series analysis to model ATM cash movement; detected seasonal patterns and trends. Applied statistical models and machine learning for demand forecasting; engineered features from transactional and seasonal cycles. Built and validated predictive models using Python (pandas, scikit-learn, statsmodels). Analyzed evaluation metrics (MAPE, RMSE, R^2) to refine forecasting accuracy. Collaborated with cross-functional teams to translate forecasts into business decisions.