PhD Position (m/f/d) – Machine Learning for Circuit Reliability Analysis

Tätigkeitsprofil:

The position:
We are seeking a highly motivated PhD student to conduct research at the intersection of machine learning, electronic design automation (EDA), and electronic circuit reliability.
The position is part of a three-year research project funded by the German Research Foundation (DFG). The project aims to develop novel machine-learning methods for the analysis and prediction of circuit reliability, with the goal of making reliability assessment more efficient, scalable, and suitable for increasingly complex electronic systems.
The successful candidate will have the opportunity to investigate fundamental research questions and develop novel approaches using modern machine-learning techniques. Depending on the research direction and the
candidate’s interests, possible approaches may include deep learning, graph neural networks, transfer learning, explainability of machine learning, and optimization methods.
The PhD student will have substantial freedom to develop their own research ideas within the framework of the project. The research is expected to lead to scientific publications at international conferences and in peerreviewed journals.

Your Responsibilities:

• Conduct independent research on machine learning for circuit reliability analysis.
• Develop novel machine-learning methods for predicting, analyzing, or optimizing circuit reliability.
• Investigate appropriate representations of electronic circuits and design data for machine-learning models
• Develop experimental frameworks and evaluate proposed methods using relevant circuit and EDA benchmarks
• Publish and present research results at international conferences and in scientific journals
• Collaborate with researchers from machine learning, circuit design, and EDA
• Contribute to the development of publications, and project activities
  

Anforderungsprofil:

Your qualifications:
Required qualifications:

• Master’s degree or equivalent in Computer Science, Electrical/Electronic Engineering, Data Science, Machine Learning, or a related field
• Strong theoretical and practical knowledge of machine learning and deep learning
• Good understanding of fundamental concepts such as optimization, generalization, and model evaluation
• Strong programming skills, particularly in Python and preferably PyTorch or a comparable machinelearning framework
• Strong analytical and problem-solving abilities
• Ability to read, understand, and critically evaluate scientific literature
• Strong interest in scientific research and the development of novel methods
• Ability and willingness to work independently and take responsibility for your own research
• Very good written and spoken English

Desirable qualifications
Experience in one or more of the following areas is advantageous:

• Electronic design automation (EDA)
• Circuit reliability, fault analysis, aging, or reliability modeling
• RTL design, Verilog/SystemVerilog, FPGA, or ASIC design
• Digital circuit design or verification
• Graph neural networks and graph representation learning
• Experience with EDA tools or circuit simulation
• Previous research experience, for example through a Master’s thesis, research project, publications, or open-source contributions

A strong background in machine learning is particularly valued. Previous experience in circuit design or EDA is an advantage but is not required, provided that the candidate has a strong interest in learning the relevant domain.

What We Are Looking For:
In addition to formal qualifications, we particularly value candidates who demonstrate:
Research independence
You can identify interesting questions, formulate hypotheses, plan experiments, and make progress without requiring detailed instructions at every step.
Scientific curiosity
You enjoy exploring new ideas and understanding why a method works, rather than simply applying existing algorithms.
Critical thinking
You can critically assess assumptions, experimental results, and existing literature, and recognize the limitations of machine-learning approaches.
Persistence
You are willing to investigate unexpected results, debug complex experiments, and revise your approach when an initial idea does not work.
Interdisciplinary interest
You are interested in understanding the underlying circuit and reliability problems and are willing to learn concepts outside your primary field of expertise.
Scientific rigor
You value reproducible experiments, meaningful baselines, appropriate evaluation metrics, ablation studies, and fair comparisons.
Initiative and ownership
You actively search for relevant literature, tools, datasets, and research opportunities and take ownership of your research direction.
Communication and collaboration
You can communicate technical ideas clearly and work effectively with researchers from different disciplines.