Research engineer position in Machine learning

Luleå University of Technology is in strong growth with world-leading competence in several research areas. We shape the future through innovative education and ground-breaking research results, and based on the Arctic region, we create global social benefit. Our scientific and artistic research and education are conducted in close collaboration with international, national and regional companies, public actors and leading universities. Luleå University of Technology has a total turnover of SEK 1.9 billion per year. We currently have 1,840 employees and 17,670 students.

In the coming years, multi-billion investments will be made in large projects in Northern Sweden to create a fossil-free society both nationally and globally. Luleå University of Technology is involved in several of these cutting-edge research projects and in the societal transformation that they entail. We offer a broad range of courses and study programmes to match the skills in demand. We hope that you will help us to build the sustainable companies and societies of the future.The division of EISLAB (Embedded Intelligent Systems LAB) targets research and innovation within the realm of the research subjects Cyber-Physical Systems, Electronic Systems and Machine Learning. We are a strong research and innovation partner in various research areas, including Electronics design and production, Electromagnetic Compatibility (EMC), IoT and SoS automation, Sensorsystems, Machine Learning and Deep Learning, Digital Transformation, and Data Science.

The machine learning group at the Department of Computer Science and Space Technology at Luleå University of Technology has notable work on using a Generative AI-driven framework for Optimal Structure Selection and Generation in Material Science. The research group is now looking for knowledgeable and committed employees to help in doing AI-driven research that simplifies the construction of machine learning force fields (MLFFs) trained on DFT results.

Project description
WASP-WISW project is using MLFFs to drive MD simulations significantly expands the length and timescales accessible compared to aiMD. Our recent work using DeepCNT-22 to drive MD simulations of SWCNT growth allowed us to model growth at microsecond timescales, enabling a detailed study of growth mechanism such as defect formation and healing.

However, challenges persist. The training of our MLFF relies heavily on state-of-the-art DFT calculations, consuming substantial time and computational resources. Our current selection of structures for the training set is suboptimal, with redundancies leading to potential biases in the MLFF.

The project goal is to develop an AI-driven framework to optimize the selection of structures for DFT labeling and to generate structures where gaps exist. This collaboration will expedite the construction of unbiased MLFFs, paving the way for studying more complex catalytic reactions. We aim to establish a protocol for efficiently training MLFFs, covering materials growth to molecular synthesis in solution.

Duties
Research and Development in AI-driven framework to optimize the selection of structures for DFT labeling and to generate structures where gaps exist. This collaboration will expedite the construction of unbiased MLFFs, paving the way for studying more complex catalytic reactions. We aim to establish a protocol for efficiently training MLFFs, covering materials growth to molecular synthesis in solution.

Qualifications
At least master’s degree in a relevant field.

Practical and theoretical knowledge in generative AI for data generation and selection.

Background in Physics is highly preferable.

Further information
Temporary position, full time. Placement: Luleå. Starting: August,2024.

For further information about the position, please contact:
Hamam Mokayed, senior lecturer, (+46) 920-49 2075 hamam.mokayed@ltu.se

Union representatives:
SACO-S Kjell Johansson, (+46)920-49 1529 kjell.johansson@ltu.se  
OFR-S Lars Frisk, (+46)920-49 1792 lars.frisk@ltu.se

In case of different interpretations of the English and Swedish versions of this announcement, the Swedish version takes precedence.

Application
We prefer that you apply for this position by clicking on the apply button below. The application should include a CV, personal letter and copies of verified diplomas from high school and universities. Mark your application with the reference number below. Your application, including diplomas, must be written in English or Swedish.

Closing date for applications: 9 May 2024
Reference number: 1108-2024