Student Assistant for the VisPer Project

The Fraunhofer Institute for Secure Information Technology SIT is one of the leading research and development institutions for cybersecurity in Germany and Europe and is part of ATHENE, the national research center for applied cybersecurity. ATHENE is a cooperation of the Fraunhofer Society with TU Darmstadt, the Darmstadt University of Applied Sciences, and Goethe University Frankfurt. Our common goal: to make the world of tomorrow safer.

Be part of change

• Implementation and application of computer vision approaches, including:
  • object detection (e.g., using YOLO, RT-DETRv2, etc.)
  • image classification using common architectures (e.g., Transformers: ViT, Swin, DeiT, CNNs: ResNet, EfficientNet, GNNs: GraphConv, GAT)
  • image description, visual question answering, and multimodal search using Vision-Language Models (VLMs) image segmentation (e.g., using SAM-3, DINO-3, etc.)
  • interactive and explainable classification systems (RISE, GradCAM, EigenCAM, Integrated Gradients, ViT Shapley, etc.)
• Implementation and execution of experiments in the field of Machine Learning (ML):
  • cleaning, preparing, splitting, visualizing data, if necessary crawling and scraping data.
  • Applying (implementation of common ML methods such as hyperparameter optimization, binary/multi-class/multi-label classification, ensemble methods, ...
  • Evaluation/benchmarking based on standardized metrics in the context of ML (e.g. accuracy, BAC, Brier, mAP, ROC, AUC, etc.) as well as functional tests.
• Development of UIs and WebApps using common frameworks (fastHTML, Streamlit, Gradio, Svelte, Flask/FastAPI)
• Participation in publicly funded and/or directly commissioned projects by industry partners

• Study of computer science, mathematics, or a related field with a focus on machine learning and ideally computer vision
• Solid knowledge in ML:
  
  • Familiar with various architectures of neural networks (especially vision transformers and CNNs)
  • Familiar with basic terms and concepts in ML (especially classification, hyperparameter optimization, fine-tuning, evaluation of models)
• Solid knowledge of Python is mandatory and will be tested in the interview
• Advantageous: ability to independently implement methods and procedures from scientific publications
• Advantageous: knowledge and experience in the field of cybersecurity
• Willingness to face new challenges
• Strong analytical thinking

• Flexible working hours that can be easily combined with your studies
• An inspiring work environment with state-of-the-art infrastructure
• The opportunity to gain practical experience and make valuable contacts in research
• The monthly working hours are 40 to 80 hours, depending on the agreement