PhD Candidate in Machine Learning for Quantum Chemistry

Job summary

This PhD position focuses on developing innovative computational methods for studying complex chemical reactions where traditional electronic-structure techniques are insufficient. The project aims to enhance automated and data-driven approaches for accurately modeling molecular systems, particularly in the context of catalysis and quantum computing applications.

Qualifications

• A two-year master's degree (120 ECTS points) or equivalent in physics, mathematics, computer science, chemistry, or a related field.
• Experience in quantum chemistry, electronic-structure theory, molecular modeling, or atomistic simulations.
• Interest in machine learning, numerical optimization, or data-driven scientific modeling.
• Programming skills, preferably in Python or a similar scientific computing language.
• Curiosity about chemical bonding, reaction mechanisms, and catalysis.
• Motivation to engage in method development rather than merely applying existing software.

Responsibilities

• Develop computational workflows for describing electronically complex chemical reactions.
• Improve molecular electronic structure representation across various reaction coordinates and environments.
• Create automated strategies for identifying relevant electronic degrees of freedom and validate them against high-level calculations.
• Utilize machine learning to enhance electronic-structure analysis and identify patterns in molecular data.
• Collaborate within an interdisciplinary research environment focused on quantum technologies and computational materials chemistry.
• Contribute to the development of software tools, reproducible workflows, and open scientific datasets.

Skills

• Strong analytical and problem-solving skills in computational chemistry and machine learning.
• Ability to work collaboratively in an international research setting.

Education

• A master's degree in a relevant field is required.

Tools

• Familiarity with quantum-chemical calculations, data-driven models, and automated workflows is beneficial.