Our Polymer Digital Engineering Lab at the University of Wisconsin–Madison is expanding! We are seeking a talented postdoctoral researcher to develop and apply machine learning interatomic potentials (MLIPs) for organic–inorganic hybrid perovskites and other molecular systems. These materials exhibit rich structural dynamics and complex chemistry across multiple length and time scales, creating exciting opportunities for predictive modeling beyond the reach of conventional atomistic simulations alone.
This role will focus on building next-generation ML-enabled atomistic models that bridge first-principles accuracy and large-scale molecular simulations. The successful candidate will work at the intersection of molecular modeling, materials design, and data-driven methods to understand structure–property relationships and accelerate the discovery of functional soft and hybrid materials.
Key Responsibilities
Develop and train machine learning interatomic potentials for hybrid perovskites and molecular systems based on first-principles data.
Perform atomistic simulations to investigate structural dynamics, phase behavior, interfaces, ion transport, and thermomechanical properties.
Construct high-quality training datasets and assess model accuracy, transferability, and uncertainty across diverse chemical environments.
Integrate MLIPs with molecular dynamics workflows to enable efficient simulation of complex materials processes beyond conventional ab initio scales.
Collaborate with experimental and computational researchers to connect simulation predictions with materials synthesis, characterization, and design.
Communicate findings through high-impact publications and presentations.
Preferred Qualifications
Education: Ph.D. in mechanical engineering, materials science, chemical engineering, chemistry, physics, or a related discipline.
Technical expertise: Strong background in atomistic simulations, electronic-structure calculations, molecular dynamics, or interatomic potential development. Experience with hybrid perovskites, molecular materials, or soft matter systems is highly desirable.
Machine-learning experience: Familiarity with machine learning for atomistic modeling, including neural network potentials, graph-based models, active learning, or related approaches.
Skills: Excellent programming and problem-solving abilities, strong written and verbal communication skills, self-motivation, and a solid publication record.
Application Process
Interested candidates should provide:
Curriculum Vitae outlining academic background, research experience, and publications.
Cover Letter detailing your interest in machine learning interatomic potentials and highlighting relevant expertise in molecular simulation, first-principles modeling, and data-driven methods.
References for at least three professional contacts.
Please send your application materials to yli2562[A.T.]wisc.edu with the subject line “Postdoctoral Application – ML Interatomic Potentials”. We will begin reviewing applications immediately and will continue until the position is filled.
Join us in developing next-generation machine learning interatomic potentials for hybrid perovskites and molecular materials, and in advancing predictive simulation tools for materials discovery and design. We look forward to welcoming a passionate researcher to our dynamic team at UW–Madison Mechanical Engineering!