Computer Science Graduate | Data Science & Machine Learning | Software Engineering 🇭🇺

I build practical data, machine learning, and software systems that turn real-world problems into useful solutions.

My work combines machine learning, data analysis, backend development, automation, and API-driven data pipelines, with a particular interest in applying ML to healthcare and other real-world problems.

• 🎓 B.Sc. in Computer Science — University of Pécs, Hungary
• 📊 Focused on Data Science, Machine Learning, and AI
• 💼 Completed internships in Data Science and Embedded Systems & Automation
• 🧠 Experience working with imbalanced datasets, model optimization, and ML evaluation
• ⚙️ Built data pipelines, REST APIs, automation workflows, and ML applications
• 🌍 Based in Hungary and open to technical opportunities across Europe
• 🚀 Interested in Data Science, Machine Learning, AI, and Software Engineering

Data Science Intern — 10Pearls Remote · 2026

• Built an end-to-end environmental data workflow for 72-hour AQI forecasting.
• Automated data ingestion, feature engineering, model retraining, and reporting.
• Developed an interactive Streamlit dashboard for communicating forecast results.
• Used Pandas, XGBoost, GitHub Actions, and Python-based data workflows.

Embedded Systems & Automation Intern — University of Pécs Pécs, Hungary · 2026

• Designed an event-driven IoT system with failure handling.
• Built a Linux-based automated file-processing pipeline.
• Troubleshot cross-system integration issues and documented solutions.
• Worked with Linux automation, SFTP/SSH, and system integration.

Programming

Data Science & Machine Learning

Backend & Development

Tools & Automation

Machine learning project focused on diabetes classification using large-scale, highly imbalanced clinical and behavioral health datasets.

• Analyzed approximately 100K clinical records and 254K BRFSS records.
• Trained and compared XGBoost, Random Forest, and Logistic Regression.
• Addressed severe class imbalance using ADASYN, class weighting, and threshold optimization.
• Achieved 0.85 recall on the clinical dataset and 0.88 recall on BRFSS.
• Focused evaluation on minority-class sensitivity, which is particularly important for screening-oriented applications.

End-to-end data science workflow for 72-hour AQI forecasting.

• Built automated data ingestion and preprocessing workflows.
• Implemented feature engineering and ML-based forecasting.
• Automated model retraining and daily reporting using GitHub Actions.
• Developed an interactive Streamlit dashboard for forecast visualization.

Java and Spring Boot backend system for university administration workflows.

• Built a RESTful backend using Java and Spring Boot.
• Implemented structured Controller, Service, and Repository layers.
• Integrated MySQL for relational data storage.
• Implemented SQL-based data handling and reporting.
• Documented APIs using Swagger/OpenAPI.

Building practical systems with data, machine learning, and software.