Postdoctoral Associate

About the role

The Postdoctoral Associate will collaborate with the Principal Investigator in the Department of Geosciences at Stony Brook University to carry out computational research on the melting behavior and physical properties of geological materials under conditions relevant to planetary interiors. This work will rely on large-scale atomistic simulations paired with machine-learning interatomic potentials.

Responsibilities

• Generate density functional theory reference data.
• Use DFT data to train and validate machine-learning force fields.
• Run molecular dynamics simulations based on those potentials.
• Analyze and interpret simulation data geophysically.
• Write up findings for peer-reviewed publication.
• Present results at conferences and group meetings.
• Participate in group activities, such as mentoring students.
• Perform other duties or projects as assigned appropriate to rank and departmental mission.

Qualifications

Required:

• Ph.D. (or foreign equivalent) in geosciences, geophysics, physics, chemistry, materials science, chemical engineering, or a closely related field, in hand by the date of appointment.
• Research experience performing first-principles electronic structure or density functional theory calculations using an established code (e.g., VASP, Quantum ESPRESSO).
• Research experience performing molecular dynamics simulations.
• Experience training, fine-tuning, or validating machine-learning interatomic potentials or force fields (e.g., MACE, NequIP, DeePMD).
• Experience running scientific calculations on Linux-based high-performance computing clusters, including scripting or workflow automation (e.g., Python, bash, or equivalent language).
• At least one peer-reviewed publication in a scientific journal.

Preferred:

• Experience with free-energy or phase-boundary calculations, such as two-phase coexistence simulations, thermodynamic integration, or the Z-method.
• Research experience with geological materials at high pressure and high temperature.
• Experience extracting transport properties from molecular dynamics trajectories.
• Experience with GPU-accelerated machine learning frameworks (e.g., CUDA, PyTorch, or GPU-enabled LAMMPS).
• Experience with an established molecular dynamics engine such as LAMMPS, ASE.
• Experience generating and curating training datasets, including active learning or uncertainty-driven sampling strategies.
• Experience presenting research at national or international scientific conferences.
• Experience preparing computing allocation requests to national facilities such as NSF ACCESS, DOE INCITE, or NERSC.
• Experience with version control and reproducible research practices, including Git and documented computational workflows.

Pay

Salary range: $58,000–$68,000 per year, plus a $4,000 location pay (annual, 1.0 FTE).

Schedule

Full-time position.