Research software engineer

The Duarte Group is a leader in computational chemistry, developing predictive tools for reaction modelling and molecular design across catalysis, sustainability, and health. We create widely used open-source software, including autodE, cgbind/C3, and mlp-train, and are pioneering new frameworks for training Machine Learning Interatomic Potentials (MLIPs). We are seeking a highly motivated Research Software Engineer to join our ERC-funded project, ML4MetaLigM, focused on advancing MLIPs for reaction modelling in the condensed phase. This role combines cutting-edge research with the development of robust, open scientific software. This post is fixed term for 3 years You will develop and implement new computational methods and MLIP workflows, contribute to high-quality, maintainable open-source code, and support reproducible research. The position includes publishing research, collaborating with national and international partners, mentoring junior researchers, and delivering training in software best practices. You will also use high-performance computing resources and contribute to the broader RSE community in Oxford.

You should hold (or be close to completing) a PhD in computational chemistry, physics, chemical engineering, computer science (chemistry-focused), or a related field. Strong programming skills in Python and C/C++ are required, along with experience in molecular simulations, machine learning, or reaction modelling. A track record of contributing to open-source scientific software using modern development practices (e.g., code review, CI) is essential, as is evidence of high-quality research outputs. Experience with ML for atomistic modelling (e.g., MACE, ACE, NequIP, PhysNet), reactive or enhanced sampling methods, community-facing open-source projects, or delivering technical training would be advantageous.