PhD Property Prediction for Embedded (AI) Systems

Recent advances in deep learning (DL) provide high accuracy for various tasks targeting a wide range of applications ranging from Tiny ML that typically run on low power devices up to foundation and large language models running on cloud-based systems. One core discipline in the development process lies the modeling and performance estimation of the target system. With this work we want to push the boundaries of the current state of the art to adapt to the increasingly rapid development cycles with new approaches to faster predict and evaluate the performance behavior of future systems.

•  In this PhD project you will investigate how to extract different hardware characteristics and the possible ways to model them at different levels of abstractions efficiency as well as accuracy.
• You will inspect how to use these characteristics to predict performance for different workloads on a target hardware platform and across platforms.
• As a part of our team you will explore different novel machine learning based methods including their applicability and efficiency compared to the conventional modeling methods.
• Furthermore you will discover different machine learning compilers and their usage as part of the modeling workflow and the related optimizations that can facilitate more efficient as well as accurate predictions.

Qualifications :

• Education: excellent degree (Master or Diploma) in Electrical Engineering Information Engineering Microelectronics or Informatics
• Experience and Knowledge: proficiency in programming languages (C/C Python Matlab) good skills in modern mathematics e.g. Machine Learning (TensorFlow PyTorch etc.) Solver Neural Networks knowledge of AI algorithms experience in digital hardware embedded systems as well as in SoC architectures
• Personality and Working Practice: you enjoy being creative and asserting yourself in certain topics; you like working in a team and understand how to think in a structured abstract and strategic way to achieve the best possible performance
• Languages: fluent in English German is an advantage

Additional Information :

March 2025

Please submit all relevant documents (CV letter of motivation certificates and links to GitHub or kaggle account).

Diversity and inclusion are not just trends for us but are firmly anchored in our corporate culture. Therefore we welcome all applications regardless of gender age disability religion ethnic origin or sexual identity.

Need support during your application
Sarah Schneck (Human Resources)
498

Need further information about the job
Falk Rehm (Functional Department)
49(172)3504799

Remote Work :

No

Employment Type :

Full-time