Research Assistant (HiWi) / Master student project at the­­­­ Max Planck Institute for Intelligent Systems in Tübingen within the Perceiving Systems Department / Software Workshop Scientific Facility.

The Max Planck Institute for Intelligent Systems in Tübingen is looking for a student research assistant/HiWi (studentische/wissenschaftliche Hilfskraft) or a student who wants to complete his/her master’s project, for working on Real time deep learning on smartphone GPU.

The scope of the project is to have a deep neural network on a mobile device that runs in real time and uses the device’s GPU. The tasks would consist in:

  1. Identify the neural network programming frameworks that run on smartphones, their advantages and limitations,
  2. Prepare a benchmark procedure and a dataset running different convolutional neural networks, and benchmark those frameworks
  3. Identify the bottleneck of each frameworks, come up with good installation practices and acceleration techniques
  4. Continue a previous work of porting an optical flow estimation onto one of those framework
  5. Propose alternative networks that may run faster on the device GPU

This pipeline will be used as a component of a bigger program, and a particular care should be taken on the software overall design. The code would preferably be written such that the NN pipeline 1) can be easily reused in other programs and 2) can easily be adapted to other NNs.

Supervision will be provided by experts in the domain of neural networks, computer vision and software development. This position is a component of a wider project, and may lead to a PhD position.

We are looking for applicants that…

  • are matriculated as master students at a German university (obligatory for HiWis) and are located in the Tübingen area,
  • have already completed a Bachelor’s degree in a field related to computer science,
  • have excellent programming skills and habits, and understand C++ well,
  • have strong backgrounds in mathematics (linear algebra, numerical methods, 3D geometry)
  • may have an programming experience with GPU and/or with smartphone (Android or iPhone)
  • have very good communication skills in English

In case of a HiWi job, the maximal number of work hours is 20 hours per week/83 hours a month, and preference would be given for applicants available for more than 40 hours per month.

What we offer

The MPI for Intelligent Systems is one of the world's leading research institutions for machine learning, computer vision and robotics, studying the concepts of intelligence and autonomy on every conceptual scale. We offer an interdisciplinary, open research environment among international staff, with connections to institutions within and outside of Germany. The post is ideally suited for students looking to extend their scientific and cultural horizons. Students without a Bachelor’s degree will earn € 9.90 per hour. Master students holding a Bachelor’s degree will be paid € 11.52 per hour.

The Max Planck Society seeks to increase the number of women in those areas where they are underrepresented and therefore explicitly encourages women to apply. The Max Planck Society is committed to increasing the number of individuals with disabilities in its workforce and therefore encourages applications from such qualified individuals.


Applications and inquiries should be sent quoting reference number 80.17 accompanied by a one-page CV to Sabrina Jung .

If you prefer to send a hardcopy application, you may do so. Please address it to:

Max Planck Institute for Intelligent Systems
Sabrina Jung
Spemannstrasse 34
72076 Tübingen

Max Planck Institute for Intelligent Systems, Tübingen site
Max-Planck-Ring 4
72076 Tübingen