Problem:
Static strategies in poker fail against adaptive opponents. The goal was to model opponent behavior dynamically and adjust decisions in real time.
Approach:
- Built an opponent modeling system using statistical analysis of player actions
- Implemented feature extraction from game history (bet sizing, frequency patterns)
- Applied machine learning techniques to classify opponent strategies
- Designed adaptive decision logic based on predicted behavior
Tech Stack:
Python, Pandas, NumPy
Outcome:
- Able to dynamically adjust strategy based on opponent tendencies
- Improved decision consistency compared to static baseline
- Modular architecture for extending with more advanced models
| Languages |
Data & ML |
Databases |
| DevOps |
Frontend |
-
🎓 MSc in Data Science (2025 – present)
Cracow University of Technology
→ Focus: Machine Learning, data analysis, optimization -
🎓 BSc in Computer Science (2020 – 2024)
Cracow University of Technology


