Caleb is an MSc student in Social Data Science studying how AI systems implicitly infer user characteristics from language and how these inferences affect model behavior. His research focuses on implicit personalization, and the ways AI systems may reproduce or amplify inequality. He is more broadly interested in alignment, AI safety, mechanistic interpretability, and fairness evaluations.
Caleb obtained a BSc in Computational Social Science from the University of Amsterdam, gaining experience collaborating with start-ups and NGOs to help address challenges in sustainability, climate change, healthcare and responsible AI through computational methods. He also worked as a research assistant under Prof. Daniel Mayerhoffer, contributing to research on intersectional perceptions of inequality using agent-based models.
AI Alignment, Sociolinguistic Bias, Implicit Personalization, Fairness and Bias Audits, Explainable AI, Intersectionality, Mechanistic Interpretability.