The Computational & Statistical Materials Science (CSMS) Lab (www.csms.io) in the Department of Materials Design and Innovation (MDI) at the University at Buffalo is inviting applications for a PhD position fully supported by a newly funded NSF DMREF project on intelligent, autonomous materials discovery.

This project will develop advanced artificial intelligence and Bayesian statistical methods to enable multi–self-driving lab (multi-SDL) data fusion, decision-making, and distributed orchestration across experimental platforms. The goal is to create an integrated scientific framework that allows multiple autonomous systems to exchange knowledge, coordinate experiments, and collectively accelerate the discovery of sustainable semiconductor and perovskite materials.

The successful candidate will work at the interface of machine learning, statistical inference, and materials science, developing algorithms that allow AI-driven experimentation to reason about uncertainty, share information across diverse data sources, and make coordinated decisions under limited resources. The work will combine theory development, computational implementation, and close collaboration with experimental partners.

Research themes include:

• Modern methods for bayesian modeling and uncertainty quantification for experimental decision-making
• Goal-oriented, reinforcement learning approaches for multi-agent orchestration under uncertainty
• Cross-platform data fusion across self-driving laboratories
• AI-assisted synthesis and characterization of sustainable semiconductor materials

Preferred background:

• Strong quantitative foundation in statistics, applied mathematics, or computational science
• Experience with Python and scientific computing
• Interest in Bayesian inference, probabilistic modeling, or AI-driven experimentation
• Familiarity with materials science and chemistry
• Familiarity with modern reinformcent learning approaches

The position offers: