Graduate ML Materials Modeling Intern

Los Alamos National Laboratory; Los Alamos (NM); On-site; USD 6,000 - 9,000; Full time

Benefits offered by this job

Medical and dental insurance

Parental leave

401(k) with company match

Tuition assistance

Flexible schedules

Onsite gyms

Relocation assistance

Job summary

Los Alamos National Laboratory in Los Alamos, NM invites applications for a 10–12 week Graduate Student Internship in computational material science. The role focuses on developing data generation methods for robust ML models and transferable interatomic potentials across diverse materials and conditions.

You will design and evaluate algorithms, implement them in Python or Julia, and work toward publication-grade results while collaborating with the AI4ND group in the X-Computational Physics

Qualifications

• Must be enrolled in a graduate program in a science or math field.
• Strong background in ML for materials science and interatomic potentials.
• Experience with Python or Julia and differentiable programming.

Responsibilities

• Characterize training dataset diversity using multiple metrics.
• Develop data generation algorithms to maximize diversity.
• Train ML models on generated datasets.
• Extensively validate model accuracy and transferability.
• Implement methods in Python or Julia.

Skills

ML surrogate models

Interatomic potentials for materials

Python

Julia

Differentiable programming

Communication skills

Independent & collaborative research

Education

Graduate program in Physics, Chemistry, Materials Science, Applied Math or related

GPA >= 3.2 (4.0 scale)

Tools

Density Functional Theory

Modular scientific software

Job description

Los Alamos National Laboratory in Los Alamos, NM invites applications for a 10–12 week Graduate Student Internship in computational material science. The role focuses on developing data generation methods for robust ML models and transferable interatomic potentials across diverse materials and conditions.

You will design and evaluate algorithms, implement them in Python or Julia, and work toward publication-grade results while collaborating with the AI4ND group in the X-Computational Physics