Олександр

Contact me Oleksandr Romanyshyn
[open contact info](look above in the "contact info" section) (Mobile)
Data analyst, Product analyst
[open contact info](look above in the "contact info" section)(email)
LinkedIn Summary
GitHub Data analyst with experience in Python (Pandas, NumPy) and
Kaggle SQL, focused on product metrics and user behavior analysis.
Built cohort retention models that informed marketing strategy.
Languages Looking to drive data-driven decisions in a product-oriented
team.
• English(B2)
• Ukrainian(Native) Experience
Jan. 2026 – Present | Junior data analyst – Remote Helpers
Hard skills
• Excel (VLOOKUP, pivot tables) Cohort analysis of user retention and optimization of
marketing strategy.
• SQL (JOINs, aggregation,
• Processed raw transaction data in Python/Pandas:
window functions)
deduplication, type conversion, user segmentation
• Python(Numpy, Pandas, Scipy,
by first purchase date.
sklearn)
• Optimized LTV calculations via vectorized Pandas
• Business metrics (MAU/DAU,
operations, achieving 5x performance improvement
Churn, ROI, LTV) over loop-based approach.
• Statistics (A/B testing, • Built cohort retention heatmap for visual anomaly detection
descriptive statistics, analysis in user behavior
of variance, correlations, • Conducted a cohort analysis, revealing a twofold
regression analysis) difference in retention rates between the January (36%) and
• BI (Tableau) July (18%) cohorts.
• Identified a downward trend in the retention of older
Courses customers, which allowed us to formulate hypotheses about low
• SQL – SQLAcademy traffic quality and seasonal shortages of goods.
• Data cleaning – Kaggle
Projects
• Data visualization – Kaggle
Revenue Drop Analysis — Coffee Shop | Python, Pandas,
Seaborn | GitHub
Education • Investigated revenue decline by analyzing the relationship
Ternopil Ivan Puluj National between new customer flow and daily revenue patterns.
Technical University • Cleaned and aggregated transactional data using Pandas;
Bachelor of Science, Computer handled missing entries and date parsing.
Engineering • Identified correlation between new customer acquisition
September 2021 - June 2025 drop and revenue dips across key periods.
• Visualized revenue trends and customer flow dynamics via
Seaborn; findings presented in Jupyter Notebook.