Studentische Hilfskraft im Bereich LLM Security

The Fraunhofer Society is seeking a Student Assistant in LLM Security to work on the development and implementation of procedures for the evaluation of the security of LLMs, evaluation of state-of-the-art and experimental procedures for post-hoc explainability of LLMs, and analysis of vulnerabilities in LLMs.

Requirements

• Enrolled student (Computer Science, Data Science, Mathematics, etc.)
• Very good Python knowledge
• Basic understanding of KI/ML mainly in relation to Transformers
• Experience with PyTorch and Transformer architectures is desirable
• Structured, independent work and teamwork
• Well-developed analytical thinking
• Good English skills

Benefits

• Flexible working hours that can be combined with your studies
• An inspiring working environment with modern infrastructure
• The opportunity to gain practical experience and establish valuable contacts in research
• Possibility for future Bachelor's and Master's thesis