Research Scientist at Google DeepMind/Brain for 5 years, working in the AI+Olfaction team and later the protein team led by Lucy Cowell
Conducted ML-based olfaction research: mapped odor space (2019), validated with human panels, discovered new mosquito repellents, explored links between odor and metabolic spaces, and engaged in industrial collaborations
Pioneered data-driven molecular design using deep learning for molecular representation, property prediction, few-shot generalization, similarity search, generation, and optimization
Applied Graph Neural Networks (GNNs) to molecular materials and investigated explainability methods for GNN trustworthiness
Co-organized a NeurIPS workshop to bridge computational discovery and laboratory validation
Background
Assistant professor at the ChemE department of University of Toronto, PI of the Chemical Cognition Lab
Researcher solving chemical problems using data-driven techniques
Designs, builds, and evaluates computational tools for molecular discovery across small molecules, polymers, chemical mixtures, and proteins
Focuses on bridging computational predictions with experimental validation, emphasizing interpretability and accessibility of research
Passionate about science education and public outreach