Dive into the future of cyber security at Volkswagen Infotainment! As a working student (m/w/d) in the field of AI Security, you help us understand the opportunities and risks of modern AI technologies such as Large Language Models (LLMs), AI Agents and Coding Assistants and integrate them safely into the context of software and product development. Together with experienced security experts, you analyze current threats, security mechanisms and best practices related to artificial intelligence and thus make a direct contribution to the safe development of innovative vehicle and software solutions. We offer: Working on topical topics at the interface of cyber security and artificial intelligence Insights into the development of secure software and AI systems in the automotive environment Cooperation with experienced security architect:innen und Security Engineers Flexible working hours for the optimal compatibility of studies and working students: Modern working environment and independent work in a motivated team Possibility to introduce current research results and innovative technologies directly into practice Become part of our team and actively shape the safe future of AI and software development at Volkswagen Infotainment. The hourly wage of a work student is €17.50.

Possible tasks of this role

Analysis of security risks of modern AI systems, in particular LLMs, Agentic AI and AI-based development tools Evaluation of AI security risks in the context of product and software development at Volkswagen Infotainment Investigation of current attack techniques and weaknesses of AI systems (e.g. Prompt Injection, Tool Abuse, Model Manipulation, Data Leakage) Analysis and Evaluation of OWASP Top 10 for LLMs and OWASP Agentic AI Top 10 Investigation and evaluation of security controls for AI systems, for example: Guard Rails Sandboxing Insulation techniques IAM concepts Audit Logging and Monitoring Security mechanisms for agent frameworks

Cooperation in the evaluation of security architectures for agentic AI systems Analysis of concepts related to agentic identities, delegation of permissions and safe tool use Creation of technical documentation, proof-of-concepts and security assessments Support for research on current developments in AI Security and AI Governance Development of small prototypes and automations with Python and Linux tools

Qualification requirements

Must have

Studies in IT/Cyber Security, Computer Science or comparable studies Very good technical affinity and interest in cyber security Finded knowledge of at least some of the following security topics: cryptography TLS, certificates and PKI Authentication, Authorization Identity and Access Management

First practical experience with: Large Language Models (LLMs) Machine learning or AI systems AI Agents and Agentic AI Coding Assistants (e.g. GitHub copilot, cursor or comparable tools)

Good knowledge of Python or Linux Shell Scripting Experience in handling Linux systems Independent and analytical methods Good knowledge of German or English

Remarkable

Knowledge of AI security concepts and techniques Experience with VS code or comparable development environments Experience with Git and version management Basic knowledge of CI/CD pipelines (GitHub, GitLab or comparable platforms) Understanding modern software development processes and DevOps/DevSecOps practices Interest in current topics such as AI Governance, AI Risk Management and Secure AI Engineering

Skills

Problem solving competence Communication Creativity Critical thinking Awareness of cybersecurity Information security Security architecture Internet Security Ingested systems Security AI governance and compliance AI engineering Artificial Intelligence (AI) Python Linux

Do you have any questions about severe disability or equality? Then our disabled representative is happy to be there for you: hard-disabled representative@volkswagen-infotainment.com