Internship in AI for Engineering Design (F/M/D)

Description

CSEM is advancing the use of artificial intelligence to accelerate the design of complex physical systems. Several ongoing projects apply ML-driven optimization and physics-informed modeling to engineering domains including photonic integrated circuits, electromagnetic antenna design, and photonic simulation. The “ Integrated and Wireless Systems” Business Unit, based in Neuchâtel, Switzerland is looking for a motivated intern to contribute to the AI and machine learning aspects of these efforts.

Your mission

You will support a multidisciplinary team in applying machine learning to inverse design and simulation problems. The internship sits at the intersection of AI and physics-based engineering: you will help develop, train, and benchmark ML models that optimize device geometries or accelerate physical simulations. Depending on your profile and interests, your work may focus on one or more of the following application areas:

• Inverse design of nonlinear photonic waveguides for supercontinuum generation, using surrogate-based and Bayesian optimization to explore large design spaces.
• AI-assisted antenna design, combining electromagnetic simulations with data-driven modeling to address multi-objective constraints and explore complex design spaces.
• Physics-informed neural networks (PINNs) for photonic simulation, developing neural surrogates that embed Maxwell’s equations to accelerate conventional solvers.

Your responsibilities

• Implement and benchmark ML optimization pipelines (e.g., Bayesian optimization, surrogate modeling, evolutionary algorithms) interfaced with physics simulators.
• Train and evaluate neural network models (e.g., PINNs, neural operators) on photonic or electromagnetic simulation data.
• Analyze results, compare AI-generated designs against human-designed baselines, and document findings.
• Collaborate with domain experts in photonics and electromagnetics to ensure physical consistency of ML outputs.
• Contribute to internal technical reports and, if results warrant, to conference or journal publications.

Your profile

Know-how Requirements:

• Currently enrolled in a Master’s program in electrical engineering, physics, computer science, or a related field.
• Solid foundation in machine learning and deep learning (coursework or project experience).
• Proficiency in Python and at least one deep learning framework (PyTorch or TensorFlow).
• Familiarity with Git and Linux environments.
• Good communication skills in English; French is a plus.

Preferred qualifications (one or more):

• Experience with optimization methods (Bayesian optimization, genetic algorithms, or gradient-based optimization).
• Exposure to physics-informed or scientific machine learning.
• Background in photonics, electromagnetics, or computational physics.
• Prior experience with simulation tools (e.g., COMSOL, or equivalent).

Interpersonal skills

• Curious, self-driven, and comfortable working in a multidisciplinary environment.
• Good problem-solving abilities and a hands-on, results-oriented approach.
• Strong collaboration and communication skills.