[EJV] Lead AI Security Engineer (LLM & Agentic AI) - with Joining Bonus
Core
Lead safety evaluation and security testing for Bosch's LLM and agentic AI systems, including prompt injection, jailbreaks, and tool abuse testing.
Role type
Lead AI Security Engineer (LLM & Agentic AI)
Builds
Safety evaluation practices, security testing pipelines, and threat models for AI-enabled products and internal tooling.
Domain
AI Safety & Security, Large Language Models, Agentic AI
Deliverable
production ML models
Required skills
Security testing, LLM safety research, prompt injection testing, threat modeling, Python/JavaScript coding, technical writing, AI governance frameworks (NIST, ISO/IEC 42001, EU AI Act)
Preferred skills
Model Context Protocol (MCP) experience, traditional penetration testing (OSCP), AI governance program experience, open-source contributions
Technologies
Python, JavaScript/Node, Model Context Protocol (MCP), OWASP Top 10 for LLM/Agentic Applications, NIST AI RMF, ISO/IEC 42001
Responsibilities
Design and run safety evaluations (red-teaming, jailbreak testing), design and run security testing (prompt injection, tool abuse), build threat models, verify human-in-the-loop controls, scope external vendor engagements, apply emerging frameworks, partner with engineering teams, produce compliance documentation, build test tooling, mentor engineers
Seniority
Senior, hands-on IC with leadership responsibilities
