AI is advancing chemistry by understanding core principles, predicting feasible reaction pathways, and accelerating drug discovery through mechanism-aware, integrative modeling.

At MIT, he leads a lab that fuses AI with medicinal chemistry to design and synthesize molecules with desirable properties, using models that grasp reaction mechanisms and physical laws.

His career includes a postdoc at the Broad Institute focused on sifting billions of candidates to identify small molecules, followed by a MIT appointment in 2020 and ongoing work pairing chemistry challenges with computational methods.

Two flagship models—ShEPhERD and FlowER—are used to guide drug discovery and ground predictions in chemistry intuition, with FlowER honoring mass balance and reaction feasibility.

As an MIT associate professor, he develops and deploys AI-driven models to analyze, design, and predict outcomes for small-molecule drug candidates, speeding the discovery process.

His work sits at the intersection of chemical engineering and computer science, applying machine learning and cheminformatics to plan reaction pathways, automate synthesis, and identify new drug molecules.

Notable lab projects include ShEPhERD, which assesses 3D protein interactions of drug candidates, and FlowER, which predicts products while enforcing conservation principles to boost accuracy.

The research emphasizes anchoring AI in chemical intuition and reaction mechanisms to broaden AI’s role in chemistry, enhance experimental design, and improve automation and planning.