Oskar – Python, Machine learning, SQL

The goal of this project was to prepare an advanced exploratory data analysis and recommendations for further data science and machine learning developments for the company's customer transportation data.

Oskar proficiently managed the following tasks:

• Gathered business requirements and project management;
• Conducted quality data analysis;
• Carried out exploratory data analysis;
• Reported results to the project's stakeholders;
• Managed the development of PoC machine learning models for demand forecasting and transport delay prediction.

The primary aim of this project was to establish a robust system capable of managing and moderating textual data originating from users of the mobile application.

Among others, Oskar succeeded in the following duties:

• Reported results to the project's stakeholders;
• Performed data pipeline cleaning and processing text data;
• Developed and deployed a machine learning model for topic modeling;
• Created and implemented a machine learning model for sentiment analysis.

This project, conducted for a leading airline, was centered around customer data analytics. Key components involved various machine learning models, including customer segmentation, propensity models, a real-time recommendation system for ancillary products on the website, and statistical analysis of marketing campaign data.

These models' predictive capabilities were leveraged to gain deeper insights into customer behavior, drive sales growth, and enhance marketing campaign targeting efficiency.

Oskar's implemented the following tasks:

• Developed machine learning classification models, predicting customer propensity to an ancillary product purchase;
• Created Flask app launching propensity models prediction when a customer is booking a flight on the website;
• Implemented machine learning clustering model for customer segmentation;
• Statistical analysis of customer and marketing campaign data;
• Cooperation with marketing and data engineering teams;
• Reporting results to the project's stakeholders.

The goal of the project was to develop machine learning models forecasting:

• Energy prices on the Polish Power Exchange market;
• Wind and solar power plant production;
• Trading risk.

Calculated forecasts were used for trading purposes in energy companies.

On this project, Oskar had proficiently carried out the following:

• Developed machine learning models forecasting future energy prices (linear models, random forests, SARIMAX models, neural networks: MLP, LSTM, CNN);
• Created and implemented energy price stochastic risk model (diffusion model, VaR);
• Developed a model forecasting renewables energy production (MLP neural networks);
• Managed statistical analysis of energy prices and renewable energy production data.