Postdoc: Physics-Grounded AI for Power-Grid Resilience

École polytechnique fédérale de Lausanne, EPFL; Vaud; Sur place; CHF 110 000 - 130 000; Plein temps

Résumé du poste

EPFL's Intelligent Maintenance and Operations Systems (IMOS) Lab invites applications for a Postdoctoral Researcher in Physics-Grounded Agentic AI for Power-System Resilience Across Timescales. The role is based in Lausanne, 100%, fixed-term, starting December 2026 or later.

The project blends agentic AI with physics-informed learning, graph neural networks, and multi-agent reinforcement learning to strengthen resilience across planning and operation.

Qualifications

• PhD completed or near completion in a relevant field (Electrical Engineering, Computer Science, Control, Applied Mathematics, Operations Research).
• Strong foundation in machine learning or optimization and solid programming skills.
• Excellent written and spoken English.

Responsabilités

• Conduct research on Physics-Grounded Agentic AI for power-system resilience across timescales.
• Develop methods using graph neural networks, physics-informed learning, and multi-agent reinforcement learning.
• Coordinate with IMOS Lab and collaborators; contribute to methodology and power-system applications.
• Build and evaluate AI models with uncertainty quantification and robustness to unseen conditions.

Connaissances

Agentic AI

Graph neural networks

Multi-agent reinforcement learning

Physics-informed learning

Hybrid modelling

Optimization

Programming

Formation

PhD in Electrical Engineering

PhD in Computer Science

PhD in Control

PhD in Applied Mathematics

PhD in Operations Research

Outils

Python

TensorFlow/PyTorch

MATLAB

Description du poste

EPFL's Intelligent Maintenance and Operations Systems (IMOS) Lab invites applications for a Postdoctoral Researcher in Physics-Grounded Agentic AI for Power-System Resilience Across Timescales. The role is based in Lausanne, 100%, fixed-term, starting December 2026 or later.

The project blends agentic AI with physics-informed learning, graph neural networks, and multi-agent reinforcement learning to strengthen resilience across planning and operation.