Machine Learning Computational Chemist

The company is developing innovative biotechnologies for the selective binding and recovery of metals from waste streams, including electronic waste. This is a hands-on senior position offering ownership of technical strategy, scientific quality and delivery.

You will work closely with computational scientists, wet-lab researchers, founders and external stakeholders to guide experimental priorities, solve complex modelling challenges and identify opportunities for intellectual property generation.

Key responsibilities

• Own and develop the machine-learning and molecular-modelling roadmap, including priorities, milestones, validation standards and computing requirements.

• Develop, adapt and validate machine-learning models and force fields for molecular applications.

• Lead the generation and quality assurance of computational datasets.

• Run and interpret quantum chemistry calculations and molecular dynamics simulations.

• Model metal-containing, biomolecular and coordination-chemistry-rich systems.

• Benchmark computational models against appropriate theoretical and experimental references.

• Diagnose and resolve issues involving electronic structure calculations, SCF convergence, molecular dynamics, model training, data quality and HPC environments.

• Build and maintain reliable, scalable HPC and GPU workflows across cloud and research computing infrastructure.

• Automate scientific workflows, training pipelines, experiment tracking and performance monitoring.

• Translate modelling results, uncertainty and limitations into clear recommendations for experimental colleagues.

• Contribute to technical strategy, programme planning and potential intellectual property generation.

• Provide technical leadership through method selection, code review, mentoring and scientific problem-solving.

Proven hands-on experience developing or adapting machine-learning models and force fields for molecular systems.

• Strong computational chemistry and molecular simulation experience, including quantum chemistry and molecular dynamics.

• Experience modelling metal-containing and biomolecular systems.

• Extensive, practical experience using HPC and GPU computing environments.

• Strong Python programming skills and experience automating scientific workflows.

• Experience using HPC schedulers such as Slurm or an equivalent platform.

• Experience building reproducible training or simulation pipelines.

• Strong understanding of metal coordination chemistry, including coordination motifs, oxidation states and binding geometries.

• Ability to troubleshoot complex problems independently across simulations, model training, datasets and computing infrastructure.

• Strong communication skills, particularly when explaining computational evidence, limitations and recommendations to experimental scientists and other stakeholders.

• The independence and adaptability needed to succeed in a fast-moving, early-stage environment.

• Early-stage biotechnology or startup experience.

Relevant technologies

Experience with some of the following would be particularly useful:

CP2K

ORCA

VASP

Quantum ESPRESSO

NWChem

Slurm or equivalent workload schedulers

Docker

Singularity or Apptainer

Weights & Biases

MLflow

Scientific Python and molecular simulation software

Qualifications and experience

You should have one of the following:

A relevant PhD plus at least three years of relevant industry or professional research experience; or

A relevant MSc plus at least five years of relevant professional experience.

Desirable experience

• Protein or peptide design.

• Metal-protein or metalloprotein systems.

• Thermodynamic calculations and quantitative assessment of molecular interactions.

• Model uncertainty estimation and data-quality assessment.

• Experiment tracking and performance optimisation.

• Mentoring or technical leadership.

• Collaboration with wet-lab scientists.

• Intellectual property generation or research commercialisation.

What success will look like

During the first three to six months, the successful candidate will be expected to deliver:

• Validated molecular machine-learning models benchmarked against appropriate computational and experimental references.

• Reliable predictions that directly inform experimental decisions.

• Clearly documented model accuracy, uncertainty, assumptions and limitations.

• Reproducible computational workflows with appropriate quality controls.

• A clear technical roadmap covering milestones, risks and computing requirements.

• Effective knowledge transfer between computational and experimental teams.