Open positions

The Materials Intelligence Research group of Prof. Boris Kozinsky at Harvard University is looking to hire new PhD students and postdocs or research scientists to develop and apply first principles and machine learning methods for computational materials physics and chemistry. 

Projects include:

1. Scientific software engineering of machine learning potentials for large scale molecular dynamics. We are developing equivariant neural network models NequIP and Allegro for interatomic potentials, that advance the state of the art in accuracy and data efficiency. Efforts are aimed at learning computationally lean and geometrically rich representations and designing methods for quantifying uncertainty of predictions. We are also developing Bayesian Force Fields in the FLARE framework that combines rigorous uncertainty in Gaussian process regression with active learning. Resulting models are implemented in LAMMPS and are used to perform reactive dynamics simulations of billions of atoms. Qualifications: Experience with implementation machine learning methods and development of high-performance scientific software with CPU and GPU parallelization, knowledge of LAMMPS.

2. Computational investigations of catalytic reactions and ionic transport. The aim is to use ML-accelerated large-scale dynamics simulations to investigate the evolution of surface structure in reactive and solvated environments, to discover mechanisms of heterogeneous reactions, to study ionic diffusion in complex inorganic and polymer electrolytes. Methods will combine first principles, molecular dynamics, enhanced sampling, and stochastic simulations with state of the art machine learning force field models. Qualifications: Familiarity with machine learning interatomic potentials, molecular dynamics, experience with modeling complex chemical systems, fluency in chemical kinetics and thermodynamics, understanding of dimensionality reduction and classification methods.

3. Development of machine learning for exchange-correlation functionals. Current work in the group is focused on improvements and performance optimizations for the recently developed CIDER formalism for designing non-local XC functionals, with an eye toward applying the resulting functionals to currently intractable problems in catalysis and energy storage materials. Effort is aimed at generating training sets with high-order quantum calculations and designing combined models for the exchange and correlation energy using non-local features of electronic density and orbitals. Qualifications: Experience with developing efficient numerical algorithms and modifying electronic structure DFT or quantum chemistry software (e.g. Quantum Espresso, PySCF, GPAW), fluency with electronic structure theory and quantum chemistry, experience with machine learning regression methods.

Preferred start date as soon as possible but flexible.

Documents should include a full CV, cover letter summarizing your experience, list of reference contacts, and up to 3 publications.

Submit application here.