Johannes Kruse

Applied Scientist II at Amazon

I work on machine learning and recommender systems at scale, with a focus on retrieval, ranking, exploration, and personalization.

My research focuses on responsible recommender systems, including beyond-accuracy evaluation, normative alignment, and the design of recommender systems that balance personalization with broader stakeholder objectives.

I hold a Ph.D. in Computer Science from the Technical University of Denmark, supervised by Dr. Kasper Lindskow, Assoc. Prof. Michael R. Andersen, Assoc. Prof. Jes Frellsen, and Prof. Lars Kai Hansen. I was also a visiting research scholar at UC San Diego, where I worked with Prof. Julian McAuley.

Before joining Amazon, I built and deployed the first large-scale recommender systems for Ekstra Bladet, serving millions of readers, organized the RecSys Challenge 2024, and created EB-NeRD, a large-scale dataset for news recommendation.

My work has been published at venues including ACM SIGIR, ACM RecSys, The Web Conference, Nature Machine Intelligence, and Journalism Practice.