Ondrej Mlynarcik
Data Scientist β’ Data Engineer β’ Automation Specialist
Building reliable data pipelines & applying ML to real problems
π§° Skills
Python Β· SQL Β· R Β· Google Cloud Β· Tableau
π Education
-
| MSc., Data Science |
University of Utrecht, the Netherlands (June 2023) |
πΌ Work Experience
Data Engineer @ Societe Generale (Oct 2024 β Present)
- Implemented automated data quality controls using Python and Teradata, reducing manual work by 85%.
- Managed data warehouse changes through deployments that ensured timely and reliable data for reporting and analytics.
- Designed an Oracle solution to efficiently parse 500M+ XML files, improving data accessibility for BI.
AI Automation Specialist @ Brandsom (Jan 2024 β May 2024)
- Used GPT-4 Vision to automate information extraction from product images, saving ~60 hours/month.
- Developed a tool with Python and prompt engineering to generate Amazon.nl and Bol.com descriptions from attributes, reducing manual work by ~70 hours/month.
Integration Specialist @ Brandsom (May 2022 β Dec 2023)
- Used regression models to analyze price elasticity, providing clients with data-driven insights into pricing strategies.
- Led the development of an ETL pipeline (scraping β Airflow β Cloud β Dashboard), delivering a new dashboard product for ASICS.
Data Analyst @ Gajos (Jan 2021 β Mar 2022)
- Built interactive dashboards in Tableau to visualize eshop and website performance for management.
- Developed SQL stored procedures to optimize product imports, saving 40 hours/month of manual work.
π Projects
Developed a reproducible, data-driven strategy to compute tree visibility nationwide in the Netherlands using the Viewshed algorithm.
- Generated a nationwide visibility dataset
- Analyzed inequality with the weighted GINI index
- Used spatial machine learning to explore socio-economic associations

Gender Representation in English Songs (1960-2020)
Analyzed gender representation in English song lyrics (1960β2020) with NLP and semantic analysis.
- Examined thematic patterns & word associations
- Measured emotional nuances with VAD scores
- Revealed differences in portrayal of gender over time

Developed a Selenium + BeautifulSoup scraper for nehnutelnosti.sk, the leading Slovak real estate platform.
- Automated property data extraction
- Addressed a clear market gap in Slovak real estate data
- Provides a reusable pipeline for further analysis

An interactive Tableau dashboard that visualizes the Guardian University Rankings for Edinburgh University.
- Explore trends in ranking, satisfaction, and finance
- Compare across universities in the UK
- Developed for stakeholders and prospective students

π Current Readings
Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow
by AurΓ©lien GΓ©ron