Physics-Informed Machine Learning - Postdoctoral Researcher

Description

The company is hiring a postdoctoral researcher to build and apply advanced computational methods for modeling complex physical systems, with a focus on physics-informed machine learning.

You will bring together machine learning, computational mechanics and numerical methods to build models that make scientific simulations more accurate, robust and computationally efficient. Possible work includes machine-learning models constrained by physical laws, learning constitutive relationships from experimental or simulation data, surrogate and reduced-order models, and adding data-driven methods to computational mechanics workflows.

The role sits in the Atmospheric, Earth and Energy Division, within the Physical and Life Sciences Directorate.

Responsibilities

• Create physics-informed machine-learning methods for computational mechanics and scientific simulation.
• Build physical laws, conservation principles, boundary conditions and material constraints into machine-learning models.
• Design, train, check and assess neural networks, surrogate models, reduced-order models and other scientific machine-learning approaches.
• Study computational and experimental data to pin down physical relationships, model parameters and constitutive behavior.
• Run finite-element, finite-volume or related numerical simulations in computational mechanics.
• Pursue your own complementary research and work with scientists both inside and outside the company.
• Join a multidisciplinary team to reach research goals.
• Write up results for peer-reviewed journals and present them at outside conferences, seminars and technical meetings.
• Take on other duties as assigned.

Requirements

• You must be able to obtain and maintain a U.S. DOE Q-level security clearance, which requires U.S. citizenship.
• A PhD in computational mechanics, mechanical engineering, applied mathematics, computational science, physics, materials science or a closely related field.
• Hands-on experience with physics-informed machine learning, scientific machine learning or similar data-driven modeling.
• Working knowledge of computational solid and fluid mechanics, numerical methods and contact mechanics for Lagrangian finite element methods.
• Experience running computational simulations and analyzing scientific or engineering data, including in high-performance computing environments.
• Ability to set up independent research projects, publish peer-reviewed papers and communicate technical material clearly, both in writing and in speech.
• Ability to work well in a multidisciplinary team.
• Experience with physics-informed neural networks, neural operators, differentiable programming or hybrid physics-based and data-driven models.
• Experience with deep learning, transfer learning, generative models, surrogate modeling or reduced-order modeling.
• Experience with finite-element methods, computational fluid dynamics, solid mechanics, multiphysics simulation or related frameworks.
• Experience with Python and machine-learning frameworks such as PyTorch, JAX or TensorFlow.

Conditions

Base pay is $123,048 per year.

The figure above reflects the lowest to highest pay the company in good faith expects to offer for this role at posting. Where you land depends on competencies, education, certifications, experience, skills, seniority, location, performance and business needs.

This is a 2-year postdoctoral appointment, extendable to a maximum of 3 years. It is a temporary, mentored training role for people early in their postdoctoral careers. You must have received your PhD or equivalent doctorate within 5 years of the expected start date.

Work is hybrid, combining in-person and virtual presence, with the option to work from home one or more days per week. Flexible schedules are available, depending on project needs.

This role requires a DOE Q-level clearance. Selected candidates will undergo a federal background investigation. All L and Q cleared employees are subject to random drug testing. Q-level clearance requires U.S. citizenship.

External applicants selected for this role must pass a post-offer, pre-employment drug test, which includes testing for marijuana, because federal law applies to the company as a federal contractor.

In Limited Areas you may be required to leave personal and laboratory mobile devices outside, including phones, tablets, fitness devices, wireless headphones and other Bluetooth or wireless devices. If a medical device pairs with a mobile device, the mobile device rules still apply. Sensitive Compartmented Information Facilities require separate approval. Hearing aids without wireless, or with wireless disabled, are allowed in Limited Areas, Secure Space and Transit/Buffer Space.

Your information will be kept confidential under EEO guidelines.

Benefits

• A flexible benefits package
• 401(k)
• Relocation assistance
• An education reimbursement program