About me
I am a Data Scientist with a specific focus on online marketplaces. My educational background in mathematics and engineering equips me with strong analytical skills to add value to online products across various platforms. I enjoy simplifying complex technical findings and offering practical insights to all team members. My communication style aims to bridge the gap between technical and non-technical stakeholders, ensuring that data-driven decisions are clear and actionable.
- Read my cv.
- Want to get in touch? Send me an email.
Resources
These are small repositories that I created and you might find useful.
- A powerful regression model for the Ames Housing dataset (Python Jupyter notebook). The Ames Housing is an updated version of a classical regression dataset. Given some dirty real-life data, we build a regression model to predict the closing selling prices of homes, based on their characteristics. We use state-of-the-art models for tabular data, including CatBoost and XGDBoost tree-based models, and we stack them for increased accuracy.
- Data Analyst interview test by the Wikimedia Foundation (Python Jupyter Notebook). This is a realistic data exploration and analysis task for an entry-level position. You can explore a notebook with a thorough analysis, another one with insights relevant for data analysts and engineers, and a third one with an executive summary for management.
- Using the Google Maps Directions API with Excel and VBA (Spreadsheet and VBA Code). Visual Basic for Applications code to retrieve driving distances between all pair of locations by address. The list of addresses is in a spreadsheet. You need a Google Maps API token to use the code.
- Discovering energy consumption patterns of commercial users (Python Jupyter Notebook). Data visualization techniques for time series data sampled hourly. It displays three powerful dataviz tools that can be used to summarize complex time series: quartile line plots, ghost line plots, and heatmaps.