The Paul Scherrer Institute PSI is the largest research institute for natural and engineering sciences within Switzerland. We perform cutting-edge research in the fields of matter and materials, energy and environment and human health. By performing fundamental and applied research, we work on sustainable solutions for major challenges facing society, science and economy. PSI is committed to the training of future generations. Therefore, about one quarter of our staff are post-docs, post-graduates or apprentices. Altogether, PSI employs 2100 people. The Science IT section is responsible for supporting a range of world-class facilities and experiments conducted at PSI. The section currently has expertise, and provides support in HPC and cloud computing for data intensive experiments through to experiment domain expertise and low-level algorithm development and deployment on GPU technology. We are also engaged in a number of national and international collaborations with similar institutions across the globe, all with the common goal of maximizing the use of large-scale facilities for science by the provision of cutting edge IT infrastructure and software.
Join a new collaboration between PSI, the Swiss Data Science Center and the Swiss National Supercomputing Centre to reduce high volume experimental data using machine learning. We are looking for a
In a collaboration with colleagues in Science IT, Photon Science Department PX, SDSC (https://datascience.ch/
) and CSCS (https://www.cscs.ch/
) you will play the key role in finding novel ways of applying machine learing to reduce high volume over 18GB/s experiment data. You will join the recently funded SDSC project RED-ML and benefit from collaborating with internationally recognised partners striving to meet the needs of more challenging experiment conditions and the need for greater scientific accuracy.
Your role will be centered on science use cases, the first of which will be the real time reduction of diffraction data from detectors capable of producing over 18GB/s and eventually over 46 GB/s of unprocessed data. The goal of this usecase would be the eventual reduction of raw data with the appliation of ML that will surpas current algorithms in both accuracy and speed.
Your main tasks will include:
- Lead and where necessary take responsibility for the design, implement and eventual support of the RED-ML collaboration, architecture and integration into PSI infrastructure
- Work on the RED-ML use cases and with domain experts to inform an architecture for data reduction as part of the end-to-end architecture being developed for Swiss Light Source SLS 2.0
- Adapt and implement software as required to operate in the PSI local and distributed environment
- Lead on the identification of an assessment of data reduction methodologies and datasets for training and verification
- Implement and test target architecture and confirm operations with data owners with eventual handover to pilot operations
- Work closely with colleagues of similar function at the PSI SLS 2.0 and SwissFEL along with leading national and international institutions
As either an enthusiastic and talented early career or experienced professional, you should be motivated to work at the interface between science and computing technology. You will have previous experience or a deep knowledge of one or more machine learning toolsets and their application. You have experience of software languages extensively used in science such as Python or C++ and developing software for deployment on on-premises or cloud based compute clusters and high performance file systems. You can demonstrate the ability to assimilate new ideas and turn them into practical, applied techniques and a willingness to continue to learn new developments. Given the data volumes and methodology being used, you will have experience of managing and organizing the parameters and results of large datasets.
You hold a PhD degree (or equivalent practical experience) in computer science or in a natural science with a high component of computer science or data science. A very good command of written and spoken English is a must, while knowledge of German is an advantage. If you are a strong team player with excellent communication skills and sense of responsibility, this position will offer you a great opportunity to develop your career in an exciting and highly multidisciplinary environment.
Our institution is based on an interdisciplinary, innovative and dynamic collaboration. You will profit from a systematic training on the job, in addition to personal development possibilities and our pronounced vocational training culture. If you wish to optimally combine work and family life or other personal interests, we are able to support you with our modern employment conditions and the on-site infrastructure. This is a fixed-term position.
The employment contract will be limited to the two-year duration of the RED-ML grant. The salary offer will be commensurate with the depth and breadth of your experience.
For further information, please contact Dr Alun Ashton, phone +41 56 310 44 81 or E-mail email@example.com.
Please submit your application online by 23 August 2020 for the position as a Software Scientist Machine Learning (index no. 9551-02).
Paul Scherrer Institut, Human Resources Management, Sabine Mier, 5232 Villigen PSI, Switzerland