About this job 

Are you ready to do meaningful internship work on technology that is changing how AI systems are built? Join Lightmatter’s Packaging team to build the foundation of an AI-enabled framework that lets engineers quickly explore package design options and narrow candidates before committing to detailed simulation or hardware builds. You’ll combine physics-based simulation with machine learning, including physics-informed neural networks (PINNs) and other surrogate models, to make fast early predictions that stay physically grounded. Initial use cases include narrowing thermal solutions and predicting warpage, with room to extend to other objectives. 

Your work will feed into all upcoming products, POCs, and test vehicles from day one, and your experiments will build fundamental understanding of coupled thermal-structural-optical physics. You’ll be mentored by experienced thermal and thermo-mechanical engineers and learn how first-order analysis, AI surrogates, and high-fidelity simulation complement each other in early design decisions, and how package architecture, materials, and cooling come together in leading-edge AI and HPC products.

Responsibilities:

• Build a flexible framework that links package design parameters (geometry, materials, process, thermal solution) to outputs such as thermal performance and warpage.
• Develop and evaluate PINNs, other surrogate models, and reduced-order or analytical models for fast first-order predictions on new package concepts.
• Automate FEA and CFD setup, execution, and post-processing in Python to generate parametric datasets. Train and validate models against high-fidelity simulation and hardware data, quantifying accuracy and uncertainty.
• Plan and run physical DOEs (test setup, measurement, analysis) to build understanding of coupled thermal-structural-optical physics.
• Partner with packaging, thermal, and thermo-mechanical engineers to define requirements and inputs/outputs, and adapt the framework to upcoming products.

Requirements: 

• Currently pursuing an MS or PhD in Mechanical Engineering or a related field.
• Coursework in heat transfer, fluid mechanics, FEA, CFD, numerical methods, and solid or fracture mechanics.
• Graduate research in multi-physics problems, with both modeling and experimental validation.
• Experience with thermal (Icepak, Fluent, Flotherm) and structural (Ansys, Abaqus, COMSOL) simulation tools.
• Proficiency in Python (or similar) for scripting, data processing, and simulation automation.
• Curiosity and willingness to learn from the team.

Preferred Qualifications

• Graduate research in electronic or semiconductor packaging.
• Familiarity with thermal-structural coupling, warpage, and package-level thermal management.
• Coursework in packaging, numerical methods, or machine learning.
• Experience with PINNs, scientific ML, or surrogate modeling of FEA/CFD data using PyTorch, TensorFlow, or JAX.
• Interest in photonics and co-packaged optics.