Working Student Cyber Security - Focus on AI Security (m/f/d)

Volkswagen Infotainment GmbH · Bochum

WerkstudentGerman needed: none

Still open on 1 Nov

You apply on the company's own website.

What you will do

Possible Tasks within this Role

Analyze security risks related to modern AI systems, including: Large Language Models (LLMs) Agentic AI systems AI-powered coding assistants
Large Language Models (LLMs)
Agentic AI systems
AI-powered coding assistants
Assess AI-related threats and vulnerabilities in the context of Volkswagen Infotainment's software and product development activities
Investigate emerging attack techniques against AI systems, such as: Prompt Injection Tool Abuse Data Leakage Model Manipulation Jailbreaking techniques
Prompt Injection
Tool Abuse
Data Leakage
Model Manipulation
Jailbreaking techniques
Study and evaluate the OWASP Top 10 for LLM Applications and the OWASP Agentic AI Top 10
Research and assess AI security controls, including: Guardrails Sandboxing Isolation mechanisms Identity and Access Management (IAM) Audit Logging and Monitoring Secure tool integration
Guardrails
Sandboxing
Isolation mechanisms
Identity and Access Management (IAM)
Audit Logging and Monitoring
Secure tool integration
Contribute to the security evaluation of agent-based AI architectures
Analyze concepts related to agent identities, delegated permissions, and secure authorization models
Build small proof-of-concepts, prototypes, and automation scripts
Support the creation of technical documentation, security assessments, and research reports
Monitor and evaluate current developments, trends, and best practices in AI Security and AI Governance

Mandatory Qualifications

Strong interest in Cyber Security and emerging AI technologies
Solid understanding of security fundamentals, especially: Cryptography TLS, Certificates and PKI Authentication and Authorization Identity and Access Management
Cryptography
TLS, Certificates and PKI
Authentication and Authorization
Identity and Access Management
Practical experience with one or more of the following: Machine Learning Large Language Models (LLMs) Agentic AI systems AI-powered coding assistants (e.g., GitHub Copilot, Cursor, Windsurf)
Machine Learning
Large Language Models (LLMs)
Agentic AI systems
AI-powered coding assistants (e.g., GitHub Copilot, Cursor, Windsurf)
Experience with scripting using Python and/or Linux Shell
Good Linux administration and troubleshooting skills
Analytical mindset and ability to work independently
Good communication skills in English

Preferred Qualifications

Knowledge of AI Security concepts, threats, and mitigation techniques
Experience with modern development environments such as Visual Studio Code
Familiarity with version control systems (Git)
Basic understanding of CI/CD pipelines (GitHub, GitLab, or comparable platforms)
Understanding of modern software development and DevSecOps practices
Interest in AI Risk Management, AI Governance, and Secure AI Engineering
Ability to investigate and understand technical security issues independently

Skills

Possible Tasks within this Role

Analyze security risks related to modern AI systems, including: Large Language Models (LLMs) Agentic AI systems AI-powered coding assistants
Large Language Models (LLMs)
Agentic AI systems
AI-powered coding assistants
Assess AI-related threats and vulnerabilities in the context of Volkswagen Infotainment's software and product development activities
Investigate emerging attack techniques against AI systems, such as: Prompt Injection Tool Abuse Data Leakage Model Manipulation Jailbreaking techniques
Prompt Injection
Tool Abuse
Data Leakage
Model Manipulation
Jailbreaking techniques
Study and evaluate the OWASP Top 10 for LLM Applications and the OWASP Agentic AI Top 10
Research and assess AI security controls, including: Guardrails Sandboxing Isolation mechanisms Identity and Access Management (IAM) Audit Logging and Monitoring Secure tool integration
Guardrails
Sandboxing
Isolation mechanisms
Identity and Access Management (IAM)
Audit Logging and Monitoring
Secure tool integration
Contribute to the security evaluation of agent-based AI architectures
Analyze concepts related to agent identities, delegated permissions, and secure authorization models
Build small proof-of-concepts, prototypes, and automation scripts
Support the creation of technical documentation, security assessments, and research reports
Monitor and evaluate current developments, trends, and best practices in AI Security and AI Governance

Mandatory Qualifications

Strong interest in Cyber Security and emerging AI technologies
Solid understanding of security fundamentals, especially: Cryptography TLS, Certificates and PKI Authentication and Authorization Identity and Access Management
Cryptography
TLS, Certificates and PKI
Authentication and Authorization
Identity and Access Management
Practical experience with one or more of the following: Machine Learning Large Language Models (LLMs) Agentic AI systems AI-powered coding assistants (e.g., GitHub Copilot, Cursor, Windsurf)
Machine Learning
Large Language Models (LLMs)
Agentic AI systems
AI-powered coding assistants (e.g., GitHub Copilot, Cursor, Windsurf)
Experience with scripting using Python and/or Linux Shell
Good Linux administration and troubleshooting skills
Analytical mindset and ability to work independently
Good communication skills in English

Preferred Qualifications

Knowledge of AI Security concepts, threats, and mitigation techniques
Experience with modern development environments such as Visual Studio Code
Familiarity with version control systems (Git)
Basic understanding of CI/CD pipelines (GitHub, GitLab, or comparable platforms)
Understanding of modern software development and DevSecOps practices
Interest in AI Risk Management, AI Governance, and Secure AI Engineering
Ability to investigate and understand technical security issues independently

Skills