Research fellow for neural networks in embedded systems

One task of our Broadband and Broadcast Department is driving the development of new technologies for edge AI and communication systems. We focus on the energy-efficient implementation of neural networks, using both state-of-the-art hardware and highly specialized accelerator architectures (neuromorphic hardware). Alongside our current research activities, we would also like to further expand our contact with industry and, in turn, incorporate our concepts into relevant applications.

Your responsibilities:

  • Conduct research into the optimization and mapping of neural networks on energy-efficient embedded systems (low-power microcontrollers, FPGAs, or special AI/neuromorphic hardware developed at Fraunhofer IIS) – always keeping the latest scientific findings in mind.
  • Develop and optimize methods and automation tools for hardware-aware neural network training.
  • Contribute to the development and validation of architectures for neuromorphic AI accelerators (including SNN accelerators).
  • Be actively involved in the design of project outlines for funding and industry projects.

What we expect from you

  • A degree with a focus on machine learning, deep learning, artificial intelligence or neuroinformatics
  • Good knowledge of the properties and application of different classes and architectures of neural networks (DNNs, RNNs, GRUs, ResNets, etc.)
  • Some initial (including practical) experience with tools for deep compression methods (e.g. quantization and pruning) and deep compilers for embedded applications (microcontrollers or FPGAs)
  • Experience in the field of learning methods for neural networks
  • You recognize the importance of incorporating the latest scientific findings into your search for effective solutions. You can evaluate new ideas independently and have a structured way of working.
  • Language skills: Very good English and good German or willingness to learn German

What you can expect from us

  • By working on a wide range of projects with a strong practical component in the field of embedded AI, you will help shape pioneering edge AI technology and neuromorphic hardware.
  • We offer an open and friendly working environment and a personalized development plan tailored to your needs through our comprehensive range of further training/education opportunities.
  • We promote your work-life balance by offering flexible working hours and various support offers that help in your balancing private life and career.
  • Our facilities and interdisciplinary collaboration enable innovative scientific work with highly sophisticated technical equipment.
  • We offer you a broad range of networking opportunities within the Fraunhofer Institute and access to active contact with several universities.
  • If you are interested and ready to take the initiative, we also offer you the opportunity to carry out a doctoral project for which we will gladly provide support.
Appointment, remuneration and social security benefits based on the public-sector collective wage agreement (TVöD). Additionally Fraunhofer may grant performance-based variable remuneration components.
In case of identical qualifications preference will be given to severely disabled candidates.
We would like to point out that the chosen job title also includes the third gender.
The Fraunhofer-Gesellschaft emphasises gender-independent professional equality.
This vacancy is also available on a part-time basis.

Fraunhofer is Europe’s largest application-oriented research organization. Our research efforts are geared entirely to people’s needs: health, security, communication, energy and the environment. As a result, the work undertaken by our researchers and developers has a significant impact on people’s lives. We are creative. We shape technology. We design products. We improve methods and techniques. We open up new vistas.

Interested? We are looking forward to receiving your application via our online application system. You may adress it to Meike Hillenbrand.

Job Reference: IIS-2021-59 Closing Date: none
counter-image