Machine Learning Materials Researcher
Job Description
Develop predictive models and reproducible workflows for computational materials research.
What you will do
• Apply machine learning to material discovery, property prediction and structure–property questions.
• Train and evaluate models using DFT, molecular-dynamics or Monte Carlo outputs.
• Use supervised and unsupervised approaches and integrate them into high-throughput materials pipelines.
• Explain modeling assumptions and document reproducible methods.
Required background
• At least one first-author publication in a peer-reviewed journal.
• Expertise in both ML and computational materials science, such as DFT, force fields, atomistic simulation or materials informatics.
• A graduate qualification at master’s or doctoral level in materials science, computational chemistry, physics, computer science or another relevant quantitative subject.
• Independent technical problem-solving ability.
Preferred background
• Teaching or TA work in computational physics, chemistry or materials science.
• CGCNN, MatGL, M3GNet, ALIGNN or related graph-network experience.
• VASP, Quantum ESPRESSO, LAMMPS, ASE or comparable materials tools.
Availability
• Remote and asynchronous; plan around 10–20 hours weekly for two to three weeks, with the scope and dates agreed before work.
Pay and engagement
Per project; fee agreed before work
The agreed project fee determines payment. Any hourly equivalent is an estimate, not an hourly-pay commitment.
Amounts are in USD. The applicable rate or fee and work requirements are agreed before you begin.