Ph.D. student
Type of contract: Temporary
Job status: Part-time
Starting date: October 2024
The Department of Computational Chemistry of the J. Heyrovský Institute of Physical Chemistry in Prague offers a part-time position for Ph.D. student focusing on the development of machine-learning potentials for molecular simulations.
Description
Machine learning potentials have become a crucial part of the set of tools used for molecular simulations. Their lower computational cost makes them a promising replacement for ab initio techniques, especially in the condensed phase. Each such model must be trained based on reference ab initio data, with the selection of the reference geometries being an especially important part of the process. This project requires the development and parametrization of such models for organic compounds solvated in water.
Requirements
- Good knowledge of machine learning principles and techniques
- Experience with the construction of equivariant message passing machine learning potentials
- Experience with training set construction and active learning techniques, including query by committee
- Experience with running parallel and GPU HPC jobs
- Ability to perform electronic structure calculations in the condensed phase, including ab initio molecular dynamics
- Python programming language, including data analysis and plotting
- Experience with: CP2K and Nequip or MACE
To apply, please send your CV, a motivation letter, and contacts for two references, including the following text in the subject line of your email: SC2024_30.
By providing us with your personal data in your CV, you have allowed J. Heyrovský Institute of Physical Chemistry to process and file your CV in the HR database. Your personal data will be used for the internal HR needs of the J. Heyrovský Institute of Physical Chemistry only. At the same time, you confirm herewith that the personal data as stated in the CV you have sent to us is true and accurate.
