Machine Learning Engineer

Optimal Inc.; Warren (MI); On-site; USD 120,000 - 180,000; Full time

Job summary

Optimal Inc. is seeking a Machine Learning Engineer to build AI models that blend physical principles with multimodal data for simulating complex systems and forecasting real-world behavior.

The role focuses on developing PINNs and physics-informed models, and on integrating diverse data types into robust predictive simulations. You will work on world-model development, combine physics-based simulations with data-driven approaches, and validate models for physical consistency using both real and

Qualifications

• Master’s or PhD in Computer Science, Machine Learning, Applied Mathematics, Physics, Engineering, or a related field.
• Hands-on experience with PINNs or scientific machine learning.
• Experience in multimodal modeling and/or world model development.
• Strong Python programming skills, with experience using PyTorch or TensorFlow.
• Knowledge of neural networks, physical systems, numerical methods, or simulation.

Responsibilities

• Develop and implement PINNs and physics-informed machine learning models.
• Build multimodal models that integrate data types including images, sensor measurements, time-series inputs, and simulation outputs.
• Develop predictive world models to learn system dynamics, simulate environments, and forecast future states.
• Combine physics-based simulations with data-driven machine learning techniques.
• Validate model accuracy, physical consistency, and performance using real-world or simulated datasets.

Skills

PINNs

Multimodal modeling

World model development

Neural networks

Physical systems

Numerical methods

Simulation

Python

Education

Master’s or PhD in a related field

Tools

PyTorch

TensorFlow

Job description

Optimal Inc. is hiring a Machine Learning Engineer to build AI models that blend physical principles with multimodal data to simulate complex systems and forecast real-world behavior.

Responsibilities

• Develop and implement Physics-Informed Neural Networks (PINNs) and physics-informed machine learning models.
• Build multimodal models that integrate data types including images, sensor measurements, time-series inputs, and simulation outputs.
• Develop predictive world models to learn system dynamics, simulate environments, and forecast future states.
• Combine physics-based simulations with data-driven machine learning techniques.
• Validate model accuracy, physical consistency, and performance using real-world or simulated datasets.

Requirements

• Master’s or PhD in Computer Science, Machine Learning, Applied Mathematics, Physics, Engineering, or a related field.
• Hands-on experience with PINNs or scientific machine learning.
• Experience in multimodal modeling and/or world model development.
• Strong Python programming skills, with experience using PyTorch or TensorFlow.
• Knowledge of neural networks, physical systems, numerical methods, or simulation.

Technologies

• Python
• PyTorch
• TensorFlow

Preferred Qualifications

• Research or industry experience in automotive, robotics, autonomous systems, or physical AI.
• Experience with sensor fusion, differential equations, simulation, or predictive dynamics models.
• Relevant research publications or project experience.

Location & Work Mode

• Warren, MI (onsite)