AI Safety & Governance Engineer

About the role

The AI Safety & Governance Engineer will join the SAGE team at NTU to ensure large language model (LLM) and generative AI systems are safe, secure, and compliant before delivery to project sponsors. You will work across diverse industry projects, mentor apprentices, and help shape Singapore's AI safety standards.

Key responsibilities

• Build and run automated LLM safety testing pipelines integrated with GitLab CI/CD as a mandatory delivery gate.
• Conduct red‑team and adversarial testing, including prompt injection, jailbreaking, data leakage, and RAG poisoning.
• Review LLM system architectures (agentic, multimodal, RAG) for security vulnerabilities and provide design recommendations.
• Document vulnerabilities with severity ratings, reproduction steps, and remediation guidance.
• Support compliance with AI safety standards such as ISO 42001, NIST AI RMF, OWASP LLM Top 10, MITRE ATLAS, and Singapore’s Model AI Governance Framework.
• Train and mentor AIAP apprentices on responsible AI engineering.

Required profile

• Degree with 2‑4 years of relevant work experience (senior level possible with strong background).
• Demonstrated ability to assess security of LLM/GenAI systems and perform adversarial testing.
• Experience reviewing system designs and recommending secure patterns for agentic, multimodal, and RAG applications.
• Strong technical writing skills for vulnerability reports and compliance evidence.

Required skills

• Python programming and testing frameworks (pytest, unittest).
• Git, Docker, and CI/CD pipelines (GitLab or GitHub).
• Working knowledge of LLM architectures, prompt engineering, Retrieval‑Augmented Generation (RAG), and API integration.
• Security testing techniques: vulnerability assessment, red‑team, adversarial testing.
• Familiarity with AI safety standards (ISO 42001, NIST AI RMF, OWASP LLM Top 10, MITRE ATLAS).