Lead AI Security Engineer (LLM & Agentic AI)

About the role

Bosch Global Software Technologies Company Limited (BGSV), Bosch's first software development centre in Southeast Asia, is looking for a Lead AI Security Engineer to own safety evaluation and security testing across its growing portfolio of AI-enabled products and internal tooling. The role covers large language model deployments and agentic AI systems, including Model Context Protocol-style tool-calling integrations, and is a cross-business-unit mandate built largely from the ground up.

Key responsibilities

• Design and run safety evaluations for LLM features and AI agents: red-teaming for harmful, biased or unsafe outputs, jailbreak and guardrail-bypass testing, and benchmarking against known safety failure modes.
• Design and run security testing: prompt injection (direct and indirect), excessive agency, tool/plugin abuse and confused-deputy risks in agentic and MCP-style architectures.
• Build and maintain a threat model covering the path from user input through model reasoning to tool execution.
• Verify that human-in-the-loop and guardrail controls are sound wherever an AI agent can take consequential actions.
• Scope, brief and review external red-teaming and pentest vendors, and write test specifications and acceptance criteria.
• Track and apply emerging frameworks such as OWASP Top 10 for LLM and Agentic Applications, NIST AI Risk Management Framework, ISO/IEC 42001 and the EU AI Act.
• Partner with AI/ML engineering teams to embed safety and security reviews early, and produce documentation for AI governance and compliance.
• Build test tooling such as adversarial-prompt harnesses, automated red-team pipelines and MCP/tool-schema fuzzers, and mentor engineers.

Required profile

• Bachelor's or Master's degree in Computer Science, Machine Learning or a related field.
• 6+ years in security testing and 2+ years in AI/ML safety research or a closely related discipline.
• 2+ years of practical experience testing or red-teaming LLM/agentic AI systems; open-source contributions accepted.
• Preferred: direct MCP experience, traditional penetration testing background (OSCP or equivalent), and experience mentoring a small safety/testing team.

Required skills

• Working knowledge of at least one AI safety/governance framework (NIST AI RMF, ISO/IEC 42001, EU AI Act).
• Coding ability in Python and/or JavaScript/Node to build evaluation harnesses and test tooling.
• Strong technical writing.
• Relevant certifications such as OSCP, ISC2 AI Security Certificate, CAISP or ISO/IEC 42001 Lead Auditor are a plus.

What we offer

• Joining bonus of 1 month gross salary by 31 December 2026, subject to eligibility.
• 13th-month salary, performance bonuses, annual appraisal, and 100% salary and social insurance during the 2-month probation.
• 15+ days of annual leave plus birthday leave, premium health insurance for employee and two family members, and flexible working hours.
• Training programs, onsite short- and long-term assignments, and company activities.