The Work
You will lead and mentor a pod of approximately 5 to 10 technical trainers who create realistic, terminal-based scientific tasks for AI training. You will review their work for scientific accuracy, reproducibility, quality, productivity, and delivery standards.
The work combines advanced materials science judgment, technical quality control, scientific software validation, team mentorship, and documentation. You will provide actionable feedback, track revisions, onboard trainers, allocate work, escalate blockers, and maintain technical documentation.
- Review problem statements, crystal structures, interatomic potentials, thermodynamic databases, simulation inputs, reference solutions, and automated tests.
- Validate tasks involving density functional theory, molecular dynamics, computational thermodynamics, phase diagrams, crystallography, materials characterization, mechanical properties, semiconductors, and energy materials.
- Check units, crystal symmetry, convergence criteria, electronic structure calculations, thermodynamic consistency, simulation parameters, dependencies, and reproducibility.
- Review automated graders for outputs such as lattice parameters, formation energies, band gaps, elastic constants, phonon properties, phase fractions, diffusion coefficients, and microstructure metrics.
- Identify inaccurate assumptions, numerical tolerance issues, scientific shortcuts, and ways required calculations could be bypassed.
What It Pays and Takes
The source listing does not provide a pay rate. This is a contractor assignment marked as part time, with an expected duration of four weeks.
- Time requirement: 20 or more hours per week.
- Location: Open worldwide.
- Language: Professional fluency in English.
- Experience level: Entry level.
- Education: Ph.D., postdoctoral experience, or equivalent advanced technical experience in materials science, metallurgy, or a closely related discipline.
- Programming: Strong skills in Python, C, C++, Fortran, or Julia.
- Technical environment: Proficiency with Linux and terminal-based environments.
- Scientific expertise: Hands-on experience with atomistic simulation, computational thermodynamics, mesoscale modeling, electronic structure calculations, or materials characterization data analysis.
- Validation skills: Strong knowledge of materials science fundamentals, numerical methods, computational modeling, scientific assumptions, numerical accuracy, simulation convergence, and computational reliability.
- Review or leadership experience: Experience reviewing scientific code, computational simulations, manuscripts, or research outputs, plus experience mentoring, reviewing, or leading small technical or research teams.
Helpful Background
Experience with scientific materials tools is helpful. Relevant tools include Quantum ESPRESSO, VASP, GPAW, ABINIT, LAMMPS, pymatgen, ASE, spglib, phonopy, pycalphad, MOOSE, FiPy, OVITO, and DREAM3D.
- Machine-learned interatomic potentials or high-throughput simulations.
- Materials databases, Docker, Git, CI/CD, automated testing, or high-performance computing environments.
- Scientific benchmarks, simulation-based evaluations, or AI coding agents.
- Background in semiconductors, batteries, alloys, advanced materials, manufacturing, publications, open-source development, or AI evaluation programs.
How It Works
Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI Training Work
AI training is the human work behind systems that learn from examples, including reviewing code, testing model outputs, and checking whether technical answers are accurate and reproducible. OpenTrain helps people find these roles, build an AI training profile, and apply in minutes, while specialist knowledge is valuable because it helps produce reliable training data.