We are looking for a post-doctoral researcher to join our research team for a project focused on enhancing the performance of hydrogen safety sensors via artificial intelligence, with the ultimate goal of implementing this approach in an industry-level sensor.

Information about the project
The main goal of this project is to radically enhance the performance of optical hydrogen sensors to achieve crucial performance improvements. This will make a key contribution to the rapid large-scale implementation of hydrogen (energy) technologies and thereby a drastic reduction in carbon dioxide emissions. This will be achieved by tightly integrating experiment and computation, developing sensing devices and analysis software in tandem. Specifically, you will develop neural network models based on recurrent and convolutional network architectures to accelerate sensing, to improve long-term stability and to enable operation in the presence of pollutants.

You will join the groups of Professor Paul Erhart (Chalmers University of Technology, Department of Physics, where you will be formally hosted; https://materialsmodeling.org/) and Professor Giovanni Volpe (Physics, Gothenburg University, Department of Physics) and collaborate with the experimental group of Professor Christoph Langhammer (Chalmers) as well as industry partners at Insplorion AB, who are specialists in sensor technology. We offer a dynamic working environment with many opportunities to expand your scientific and technical skill set, as the host environment comprises an international group of researchers and engineers with extensive expertise in method development, computational techniques, and sensor technology. In addition there are various opportunities for collaboration with local and international experts in experiment and computation in both academia and industry. 

Overall this project will enable you to deepen your understanding of machine learning as well as its applications in the natural sciences, and enable you to become proficient and/or upgrade your skills in computational methods, including machine learning techniques, in application-oriented research. It will also allow you to expand your knowledge of state-of-the-art software and data engineering techniques in an academic environment with strong ties to the industry. Moreover, you will be able to work in a cross-disciplinary, collaborative environment and have ample opportunities to grow your professional network, e.g., via collaboration, conferences and workshops.

Major responsibilities
The project comprises both a computational and an experimental post-doctoral researcher. The computational position announced here is focus on the development of models for accelerated sensing and for improving stability over time as well as in the presence of pollutants. To address these aspects you will build recurrent and convolutional neural network models, using both synthetic and actual data produced in the experimental part of this project. As a part of this effort you will participate in the design of a sensor with properties optimized for the use in the neural network models. To enable the integration of these models in real devices, you will also work on delivering these models as software.

In this project, you are expected to plan, conduct and analyze numerical simulations, design and train neural network models, to interpret the results and to drive the progress of your project. We also encourage you to supervise Master and doctoral students and will support you in this process.      

Contract terms
Full-time temporary employment. The position is limited to a maximum of two years (1+1).

Qualifications
To qualify for the position of postdoc, you must have a doctoral degree in a relevant field; Physics, Electrical engineering, Computer science, Chemistry or a related discipline. You should also have a strong interest in machine learning in the context of the natural sciences with demonstrated expertise in at least two of the following areas:
• Machine learning, in particular neural networks
• optical sensing
• software engineering
• graphical user interfaces

As many of our workflows are based on Python, good command of this tool is mandatory. In this context, prior experience with Tensorflow and/or PyTorch is very beneficial. You should enjoy working in a collaborative environment including interactions with both computational and experimental researchers and engineers.

You are expected to be somewhat accustomed to teaching, and to demonstrate good potential within research and education. The position requires sound verbal and written communication skills in English. Swedish is not a requirement but Chalmers offers Swedish courses.

Please use the link below to apply:
https://www.chalmers.se/en/about-chalmers/Working-at-Chalmers/Vacancies/Pages/default.aspx?rmpage=job&rmjob=9961&rmlang=UK