This PhD position explores how AI agents can play games to generate meaningful gameplay data. You will work on reinforcement learning, automated feature engineering, and the comparison of AI- and human-generated gameplay.
Background: Video games shape behaviour, well-being, and digital economies—yet we lack scalable, objective ways to analyse
gameplay itself. In this PhD project, you will tackle the scientific challenge of developing AI agents that can play games, generate gameplay data at scale, and automatically extract meaningful gameplay features linked to human experience.
Your Role: You will design, train, and evaluate reinforcement-learning agents that play classic and modern video games. Your work will focus on automated gameplay analysis, feature engineering from large-scale behavioural data, and comparing AI-generated gameplay with human play. You will contribute core methods that enable scalable, explainable, and human-centred game AI.
Context: You will contribute to the
ERC Starting Grant GAMECHAR (Scalable AI-Driven Framework for Gameplay Characterization;
https://cordis.europa.eu/project/id/101220528) within the
Human-Technology Interaction (HTI) group at Eindhoven University of Technology.
HTI brings together psychology, AI, data science, and design to study how interactive systems shape human behaviour and experience. You will collaborate closely with another PhD researcher and a postdoctoral researcher, and work with national and international partners.
https://www.tue.nl/en/research/research-groups/innovation-sciences/human-technology-interactionThe position is embedded in TU/e’s strong AI ecosystem and connected to the
Eindhoven AI Systems Institute (EAISI) and game research facilities.
Societal Impact: Your research contributes to responsible and human-centred AI. By enabling objective gameplay analysis, your work supports better digital health interventions, fairer game design, and evidence-informed European policy on deceptive and harmful digital practices.