Industry Keynote:  Annalisa Pawlosky & Petar Sirkovic (Google)

Title: Building Self-Play Labs2 for AI scientists: A Focus on ALS

Annalisa Pawlosky PhD, MBA, is a Senior Staff Research Scientist at Google with 20 years of experience building mathematical models and synergistic experimental techniques. She is a lead for building self play labs for AI co-scientists with labs located in San Francisco, Zürich and Singapore. Her research has a particular eye on areas such as ALS, cancer, protein engineering, and IDPs (Intrinsically Disordered Proteins). Prior to her work at Google, she completed a postdoc at Stanford University and a PhD at Harvard/MIT in Health Sciences and Technology (HST).

Petar Sirkovic is Senior Staff Software Engineer at Google DeepMind who has been working on mathematical modelling for 15 years, and has been working with GenAI and agentic capabilities since their inception. He was the quality tech lead for the AI Coscientist project and has also worked on agentic lab automation.

His particular passion is using novel GenAI techniques for pushing the scientific frontier and improving the people's lives. Before joining Google, Petar completed his PhD at EPFL and worked on an F1 team.

Academia Keynote:  Catherine Jutzeler (ETH Zurich)

Title: Early Detection of Sepsis Under Label Uncertainty

Abstract: We analyse data from approximately 20,000 patients across five Swiss hospitals to investigate early sepsis detection in real-world clinical data. A major challenge is uncertainty in the sepsis definition and resulting labels, which can substantially affect estimates of sepsis prevalence and the performance of predictive models. In this talk, we demonstrate how different approaches to defining sepsis influence observed prevalence and discuss the implications of label uncertainty for the development and evaluation of early warning models.

Prof. Dr. Catherine Jutzeler is an Assistant Professor of Biomedical Data Science at ETH Zurich and heads the Biomedical Data Science Lab at the Department of Health Sciences and Technology. She leads NDS-IICU, a Swiss-wide multicentre study investigating sepsis and critical illness using large-scale clinical, microbiological, and genomic data. Her research focuses on applying machine learning and data science to real-world clinical data to improve disease prediction, biomarker discovery, and precision medicine. Her interdisciplinary work spans sepsis, spinal cord injury, oncology, lower back pain, and neurological diseases, with a strong emphasis on translating data-driven methods into clinically meaningful applications.