About
I am the Carl and Amy Jones Professor in Interdisciplinary Science at Carnegie Mellon University, with appointments in the Department of Chemistry and Materials Science and Engineering. I lead a research group developing machine learning methods for computational chemistry and drug discovery. My work focuses on creating transferable neural network potentials that bridge quantum mechanical accuracy with computational efficiency, enabling molecular simulations at unprecedented scale.
My research spans machine learning interatomic potentials, generative molecular design, reaction prediction, and the integration of computational methods with automated experimentation. A central theme is the development of methods that are not merely accurate on benchmark datasets but remain reliable when deployed in novel chemical contexts—an essential requirement for genuine scientific discovery.
Prior to joining CMU, I held positions at the University of North Carolina at Chapel Hill (2013–2019) and the US Army Engineering Research & Development Center. I received my Ph.D. in Theoretical Chemistry from Jackson State University under the mentorship of Prof. Jerzy Leszczynski, followed by postdoctoral training at Case Western Reserve University with Prof. Carlos E. Crespo-Hernández.
Research Philosophy
I believe that machine learning in chemistry should be guided by physical principles rather than treated as purely statistical pattern matching. The most impactful ML methods encode known symmetries, conservation laws, and chemical intuition into their architecture—not as constraints that limit flexibility, but as inductive biases that enable generalization.
Equally important is the integration of computational predictions with experimental validation. Self-driving laboratories and closed-loop optimization represent the future of chemical discovery, where ML models not only make predictions but actively learn from experimental outcomes to improve their reliability.
Our goal is not to replace the expertise of chemists but to augment it—providing quantitative tools that enable exploration of chemical space at scales and speeds that would be impossible through intuition alone.
Education & Career
Academic Positions
Education
Selected Awards
Contact
Office
Mellon Institute, Room 511A4400 Fifth Avenue
Pittsburgh, PA 15213
Phone: 412-268-3140
Prospective Students
I am always looking for motivated graduate students and postdocs to join our research group. We seek candidates with backgrounds in:
- Machine learning and deep learning
- Computational chemistry and quantum mechanics
- Cheminformatics and drug discovery
- Scientific software development
If interested, please send your CV, transcripts, and a brief description of your research interests to olexandr@cmu.edu.