Researcher building and interrogating a new class of AI systems for direct retrosynthesis. My work accelerates the discovery of life-saving drugs and advanced materials by integrating synthetic feasibility directly into high-throughput molecular screening.
I operate on a simple principle: a discovery is only as valuable as its ability to be used, scaled, and built upon. My research focuses on creating new AI models that address core bottlenecks in science, like quantifying the synthetic feasibility of computer-generated molecules. But a model is not a tool until it is in the hands of a chemist, so I engineer and deploy my work as robust, full-stack applications. Finally, progress compounds when knowledge is made durable. Through pedagogy and system-building, I work to ensure today's frontier becomes tomorrow's reliable foundation.
Developing models that generate entire synthesis routes as coherent sequences, avoiding the cascading errors of traditional iterative methods.
Designing AI strategies that strategically explore chemical space to find novel drug candidates, validated by reproducing known FDA-approved drugs.
Creating composite methods to predict quantum properties with gold-standard accuracy at a fraction of the computational cost.
Investigating the internal reasoning of chemistry AI to enhance transparency, reliability, and the industrial trust needed for real-world adoption.
A full-stack application that hosts computational chemistry models from 5 publications, making them accessible to the public. Has served over 130 users and processed more than 1,000 prediction tasks.
A breakdown of ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generation
we built an ai steering system for molecular design that efficiently guides a generative model to invent potent, protein-specific molecules, even rediscovering existing drugs from scratch
A breakdown of MP2-Based Composite Extrapolation Schemes Can Predict Core-Ionization Energies for First-Row Elements with Coupled-Cluster Level Accuracy
we developed a computational shortcut that predicts quantum properties with gold-standard accuracy at a fraction of the cost, making high-precision modeling accessible for larger molecules.