Victor Fung is a Director of AI for Simulations at Lila Sciences, a Boston-based startup developing a scientific superintelligence platform and autonomous labs for the life and physical sciences. He is also an Assistant Professor in the School of Computational Science and Engineering. He obtained his PhD in Chemistry at the University of Riverside, California in 2019 and his BS in Chemistry at Cornell University in 2015. Previously, he was a distinguished staff fellow at Oak Ridge National Laboratory from 2019-2022. Dr. Fung’s research lies at the intersection of theoretical chemistry, materials modelling, and AI. His current research seeks to develop AI/machine learning methods and frameworks for accelerating materials discovery and materials simulations, in particular exploring how agentic AI can automate the scientific process.
Presentation Title:
Towards scientific superintelligence: autonomous agentic frameworks for accelerated materials discovery
Presentation Abstract:
At Lila Sciences, we are building scientific superintelligence — AI that can rapidly formulate hypotheses, design experiments, execute complex simulation and experimental campaigns, and reason over results to drive discovery and optimization cycles — and deploying them to perform autonomous science across a broad range of scientific domains. In this talk, we will provide a general overview of these efforts within LILA, including specific demonstrations in computational materials science and materials discovery. There, we will show how an autonomous agent orchestrates the end-to-end closed-loop discovery of magnetic materials — from literature synthesis through high-throughput ab initio and surrogate-driven simulations to Pareto-optimal candidate selection — completing in hours what would traditionally require months of expert effort.