AI/ML Scientist Lead Engineer

The Role

We're looking for a visionary Physics AI leader to drive our vision for Physics AI. This role is a player-coach who will lead the Physics AI team at Luminary, while contributing concrete ideas and product architecture to drive the delivery of Physics AI foundation models. The role is responsible for driving how Luminary changes customer engineering design workflows forever

Responsibilities

• Develop Physics-AI Tooling: Architect and implement high-performance tools for physics-informed workflows, similar in scope and capability to NVIDIA Modulus/Physics-ML (formerly Physics-Nemo), ensuring the delivery of models built off of synthetic data
• Foundation Model Research: Lead the development of large-scale foundation models for the physical sciences, inspired by the collaborative, cross-domain approach of initiatives like Polymathic AI.
• Architectural Innovation: Design and optimize specialized neural architectures for multi-scale physical systems, e.g.AB-UPT and related operator learning methods.
• Physics-Informed Machine Learning (PIML): Embed physical constraints (conservation laws, symmetries, and PDEs) directly into the loss functions and inductive biases of deep learning models to ensure physical consistency and data efficiency.
• Scalable Engineering: Collaborate with software engineers to deploy these models at scale within the Luminary Cloud platform, enabling real-time or near-real-time simulation for complex CFD/FEA problems.
• Leadership: Drive the deliverables of the physics AI team each quarter contributing to the larger Luminary platform

Qualifications

• Masters degree or higher in Computer Science, Mechanical Engineering, Aerospace Engineering, or related field
• 5+ years of experience building production software or ML systems
• Experience with Physics Nemo models such as Domino and GeoTransolver
• Experience with Geometry processing, Meshing, and physics solvers a must
• Familiarity with developing LLM-powered applications a plus
• Strong proficiency in Python
• Proficiency using coding agents such as Claude Code
• Familiarity with Physics AI, CAE, or physics simulation domains a critical requirement
• Experience with distributed ML applications a big plus

What we are not looking for

• Not looking for a pure manager for this role
• Not looking for someone who has no background in Physics