Doctoral student in physics-guided foundation model for time-series data Chalmers University of Technology in Sweden

Chalmers University of Technology is offering a fully funded Doctoral student position in physics-guided foundation models for time-series data. The project sits at the intersection of machine learning, time-series modeling, and physics-aware AI, with a primary focus on automotive safety-critical systems. The work is carried out at the Department of Computer Science and Engineering, within the Division of Computing Science, and is a collaboration between AIXLab@Chalmers and Volvo Group.

The doctoral student will develop reusable, pretrained foundation models for multivariate time-series that can be adapted across vehicles, driving conditions, and tasks. Research topics include predictive and generative modeling, representation learning, forecasting, and scenario generation under safety and reliability constraints. The project emphasizes physics guidance through system constraints, inductive biases, hybrid simulation-learning loops, and loss formulations that encode physical consistency. Data will include CAN signals, sensor streams, and simulated state trajectories, with validation in simulation and with industry partners.

This is a strong opportunity for candidates interested in combining theory with real-world applications. You will work closely with researchers and engineers at Volvo Group and have access to industrial datasets, simulation environments, and validation workflows. The project aims to improve safer automation, reduce failure modes, increase testing efficiency, and lower energy use. Although the main use case is automotive, the methods are intended to transfer to other safety-critical domains such as healthcare.

Eligibility highlights: applicants must hold a Master’s degree in Computer Science, Electrical Engineering, or equivalent; for degrees earned outside Sweden, a 4-year Bachelor’s degree is accepted. Strong English communication skills are required, along with solid foundations in machine learning, probability, statistics, and optimization. Experience with Python, PyTorch, large-scale GPU or cluster-based training, reproducible pipelines, and systematic evaluation will be important. Prior work in physics-informed ML, time-series foundation models, safety-critical systems, or academic publication is a plus.

Funding and terms: the position is fully funded from the start. It is limited to four years, with the possibility to teach up to 20%, extending the total length to five years. The starting salary is 34,550 SEK per month. Doctoral studies require physical presence for the whole study period, and a valid residence permit must be presented by the start date.

Application deadline: 1 October 2026. Applications must be submitted in English as PDF files and should include a CV, personal letter, bachelor’s thesis, and if available, master’s thesis and transcripts. Applications sent by email will not be considered.