Bio

If you need my photo/bio for an invited talk or another event, please grab it here

Bio for invited talks: Gabriele Sarti is a Postdoctoral Researcher at the Bau Lab of Northeastern University, where he works on scalable interpretability methods and interfaces for the National Deep Inference Fabric (NDIF) project. He previously completed his PhD at the University of Groningen, with an internship at Amazon Translate and a thesis on actionable interpretability which received the EAMT 2025 Best Thesis Award. His research spans various aspects of interpretability, including context attribution, user modeling, deception detection and white-box monitoring of LLM agents. His work regularly appears in top AI and NLP conferences and journals, and has been featured in Wired, Science and Die Zeit. He helped develop several popular open-source interpretability libraries such as NNsight and Inseq, co-organizes the BlackboxNLP workshop, and regularly serves as mentor in AI safety upskilling programs such as SPAR and CBAI.

Short bio: Gabriele Sarti is a Postdoctoral Researcher at the Bau Lab of Northeastern University, where he works on scalable interpretability methods and interfaces for the National Deep Inference Fabric (NDIF) project. His research spans various aspects of interpretability, including context attribution, user modeling, deception detection and white-box monitoring of LLM agents. He is a maintainer of various open-source interpretability libraries, a co-organizer of the BlackboxNLP workshop, and a regular mentor for various AI safety upskilling programs.