Machine learning for perception, interaction and reliable intelligent systems.

I am Dr Samuel Adebayo, a machine-learning researcher and computer scientist at Queen's University Belfast, where I completed my PhD in Computer Science and now work as a Research Fellow. My research began with visual perception and human-machine interaction, studying how gaze, hand motion and object interaction can support intention inference in collaborative tasks. Through ARISE, my current academic work extends this into machine learning and computer vision for robotic structural assembly within a US-Ireland collaboration involving Queen's, New York University, the University of Galway and the University of Texas at San Antonio.

My route into research has also been shaped by engineering and applied data work. Before and alongside academia, I worked across industrial engineering, automation and energy, including TotalEnergies, Procter & Gamble and Hayat Kimya. More recently, I have worked on production AI through ISx4 and Deloitte, including agentic workflows for Air Canada, with emphasis on evaluation, observability, policy-aware decision-making and human escalation. I also teach machine learning and analytics at Belfast Metropolitan College.

Across these settings, I am interested in how intelligent systems extract useful state from imperfect observations, act on that information and remain reliable enough for real-world use. That interest now spans computer vision, multimodal and temporal learning, embodied intelligence, LLMs and agentic systems.