Doctoral student in physics-guided foundation model for time-series data

Chalmers Tekniska Högskola AB; Göteborgs kommun; On-site; SEK 351,000 - 420,000; Full time

Benefits offered by this job

Employee benefits

Job summary

Chalmers Tekniska Högskola AB invites applications for a Doctoral student position in physics-guided foundation models for multivariate time-series data, with automotive applications. You will design predictive and generative models and validate them in simulation to advance safe automation.

The project is a collaboration between AIXLab@Chalmers and Volvo Group. You will work with engineers and researchers on real datasets, scaling training on GPUs and contributing to publications and industrial

Qualifications

• Master's degree (120 credits) in CS/EE or equivalent; 60 credits magisterexamen accepted.
• Strong English written and verbal communication.
• Solid ML fundamentals and interest in time-series and physics-guided ML.
• Proficiency in Python and PyTorch.
• Experience with large-scale GPU training and reproducible pipelines.
• Ability to formulate research questions and run scalable empirical studies.

Responsibilities

• Design predictive and generative models for multivariate time-series.
• Scale training on real data and validate in simulation with industry partners.
• Develop physics-guided architectures incorporating dynamics and conservation laws.
• Publish results in top-tier ML or AI venues and contribute to industrial validation pipelines.

Skills

English proficiency

Machine learning fundamentals

Python & PyTorch

Research design

Large-scale training

Time-series modeling

Education

Master's degree (120 credits) in Computer Science or Electrical Engineering

4-year Bachelor's degree accepted outside Sweden

Tools

GPU clusters

PyTorch

Job description

Doctoral student in physics-guided foundation model for time-series data

REF 2026-0434

Join us to develop physics-guided, data-driven foundation models for multivariate time-series in safety-critical systems, with a primary focus on automotive applications. You will design predictive and generative models, scale training on real data, and validate in simulation and with industry partners to advance safe, reliable automation.

About us

The Department of Computer Science and Engineering , a joint department of Chalmers and the University of Gothenburg. Our internationally visible research, strong industry links and diverse environment create a collaborative setting where ideas grow into real impact.

Atthe Division of Computing Science , we advance secure and trustworthy software and systems, spanning foundations, programming languages,toolsand practical methods that help shape dependable digital infrastructures.

This project is a collaboration between the AIXLab@Chalmers and Volvo Group. You will be joining us at the AIXLab, where we focus on developing AI solutions that are usable and applicable in real-world settings. In addition, you will work closely with engineers and researchers at Volvo Group, with direct access to industrial datasets, simulation environments, and real validation workflows.

About the research project

This project advances physics-aware foundation models for time-series data. Here, foundation models refer to reusable, pretrained time-series models that can be adapted across vehicles, driving conditions, and tasks. The primary use case is automotive: predicting vehicle behavior, simulating rare safety-critical scenarios and generating test cases to strengthen validation and reduce physical trials. The techniques are designed to transfer to other safety-critical domains such as healthcare.

Concretely, the research will focus on multivariate vehicle time series such as CAN signals, sensor streams, and simulated state trajectories, with models that integrate physical structure (e.g. dynamics, constraints, conservation laws) into large neural architectures. Physics guidance may include explicit system constraints, inductive biases in model architectures, hybrid simulation learning loops, or loss formulations that encode physical consistency. The work combines forecasting, representation learning, and scenario generation under safety and reliability constraints.The results will support safer automation, fewer failure modes, more efficient testing, and lower energy use.

Who we are looking for

We are particularly interested in candidates who enjoy working at the intersection of theory, data, and real-world systems, and who are comfortable with imperfect, noisy, and safety-constrained data.

The following requirements are mandatory:

• To qualify as a Doctoral student, you must have a Master's degree (masterexamen) of 120 credits or a Master’s degree (magisterexamen) of 60 credits* in Computer Science, Electrical engineering, or equivalent.
• You will need strong written and verbal communication skills in English.
• Strong machine learning fundamentals (probability, statistics, optimization) and strong interest in time-series modeling and physics-guided machine learning.
• Proficiency in Python and modern deep learning frameworks (e.g., PyTorch).
• Strong engineering maturity, including experience with large-scale GPU or cluster-based training, reproducible experiment pipelines, versioned datasets, and systematic evaluation.
• Ability to formulate research questions, run empirical studies at scale.

*for students with an education earned outside of Sweden, a 4-year Bachelor’s degree is accepted.

The following experience will strengthen your application:

• Experience with physics-informed machine learning
• Background in foundation models for time-series (forecasting, representation learning)
• Exposure to safety-critical systems, scenario generation, or test coverage for edge cases
• Experience with academic research and publications
• Develop your own scientific concepts and communicate the results of your research verbally and in writing
• By the end of the PhD, the candidate is expected to have developed reusable modeling frameworks for safety-critical time-series data. The work should result in publications in top-tier machine learning or applied AI venues and contributions that influence industrial validation pipelines.

Contract terms

• The Doctoral student positions are fully funded from start.
• The position is limited to four years, with the possibility to teach up to 20%, which extends the position to five years.
• A starting salary of 34,550 SEK per month (valid from May 25, 2025).
• Doctoral studies require physical presence throughout the entire study period. A valid residence permit must be presented by the study start date; otherwise the admission may be withdrawn.

What we offer

• As a Doctoral student at Chalmers, you are an employee and enjoy all employee benefits. Read more aboutworking at Chalmers and ourbenefits for employees.
• A dynamic and inspiring working environment in the coastal city of Gothenburg .
• Read more about Sweden’s generous parental leave, subsidized day care, free schools, healthcare etc at Move To Gothenburg.

If Swedish is not your native language, Chalmers offers Swedish courses to help you settle in.

Find more general information about doctoral studies at Chalmers here .