AI Research Assistant for Power Electronics Design

Aalborg Universitet; Aalborg Kommune; On-site; DKK 420,000 - 540,000; Full time

Job summary

AAU Energy at Aalborg University invites applications for a fixed-term Research Assistant position in Artificial Intelligence for Power Electronics Design. The role starts January 1, 2027 and lasts two years, based at the Aalborg campus.

You will develop, train and evaluate machine-learning models using circuit data, board geometries, simulation results and design images. The work includes building surrogate models that accelerate solvers, applying computer vision to PCB images, and exploring

Qualifications

• Master's degree required in a relevant field.
• Solid foundation in applied machine learning.
• Experience with developing, training and evaluating ML models.
• Familiarity with electronic systems, PCBs and circuit design context.
• Experience with graph-based ML or graph neural networks is an advantage.

Responsibilities

• Develop, train and evaluate ML models based on circuit data, board geometries and design images.
• Develop surrogate models to approximate solvers for faster design evaluation.
• Explore computer vision to recognize circuit structures and PCB geometries from images.
• Investigate graph-based methods to model circuit relationships and rules.
• Contribute to AI-enabled workflows for intelligent electronic-system design.

Skills

Machine learning

Computer vision

Graph neural networks

Data-driven methods

English communication

Education

Master's degree in computer science, electrical engineering, electronics, or artificial intelligence

Job description

AAU Energy at Aalborg University invites applications for a fixed-term Research Assistant position in Artificial Intelligence for Power Electronics Design. The role starts January 1, 2027 and lasts two years, based at the Aalborg campus.

You will develop, train and evaluate machine-learning models using circuit data, board geometries, simulation results and design images. The work includes building surrogate models that accelerate solvers, applying computer vision to PCB images, and exploring