Machine Learning Engineer (Physics-based)
At a glance
Mid-level Machine Learning Engineering role at Root Access. New York · full-time · no salary stated · we estimate $166,050–$230,000.
Root Access has not published a salary for this role. The figure below is our estimate, not theirs.
aimlroles summary, based on the employer's posting.
What you'll do
- Scale automated workflows for electrical, thermal and electromagnetic simulations
- Build faster-running stand-ins for physics simulations, then check their predictions against solvers and experiments
- Turn simulation outputs and physical measurements into datasets for model training
- Use engineering constraints to shape models, train them and assess their results
- Use active learning and optimization to rank candidate designs and decide which simulations to run
What you bring
- Strong Python skills plus hands-on use of PyTorch, JAX or a similar machine-learning framework
- Experience building and assessing machine-learning models for scientific or engineering problems
- You understand numerical techniques and physics-based modeling, and have automated simulation software or computational solvers
- A degree in physics, applied mathematics, computer science, electrical engineering or a related area, or equivalent experience
Who this fits
This role suits a mid-level engineer who can connect physical modeling with reliable software pipelines and large engineering datasets. You’ll work across hardware and machine-learning teams. The job is onsite in New York.
From the employer
Role Description
Root Access is hiring a Machine Learning Engineer to develop models that predict and optimize the behavior of electronic systems. You will build automated simulation pipelines, generate training datasets, and develop models grounded in electrical, thermal, and electromagnetic physics. Working across our hardware and machine learning teams, you will turn engineering simulations and physical measurements into tools that accelerate electronics development.
Responsibilities
Build and scale automated workflows for electrical, thermal, and electromagnetic simulations.
Develop surrogate models that approximate physics simulations with faster inference.
Create training datasets from simulations and hardware measurements.
Incorporate physical constraints and engineering knowledge into model architectures, training, and evaluation.
Validate model predictions against established solvers and experimental results.
Develop active learning and optimization workflows to identify promising designs and prioritize simulations.
Own reproducible pipelines for data generation, model training, evaluation, and deployment.
Qualifications
Strong Python skills and experience with PyTorch, JAX, or similar ML frameworks.
Experience developing and evaluating machine learning models for scientific or engineering applications.
Working knowledge of numerical methods and physics-based modeling.
Experience automating simulation tools or computational solvers.
Ability to build reliable software pipelines and work with large engineering datasets.
Degree in physics, applied mathematics, computer science, electrical engineering, or a related field, or equivalent experience.
Preferred: PhD with experience in simulations, surrogate models, neural operators, or physics-informed machine learning.