Hi, I am a final-year PhD candidate at the University of Cambridge. My research focuses on making machine learning models unlearn undesired data without having to retrain the whole model. This can be used to remove privacy infringing, copyrighted, erroneous, poisoned, outdated or otherwise problematic data. Machine unlearning is not only beneficial for data removal but also helps to deepen our understanding in how/where models store information. I also work on some applications of agentic LLMs within research and hackathons.
Next to my studies, I am working part-time as a Student Researcher at Google DeepMind and have previoulsy interned at IBM Research in their AI Security & Privacy team working on automated LLM Red Teaming. Prior to my PhD I worked at the Boston Consulting Group (BCG) in Zurich and co-founded an ESA BIC-funded startup.
📧 Feel free to reach out for collaborations, talks, or career opportunities.