Senior AI Security Engineer

As a Senior AI Security Engineer, you will play a hands‑on role in building and embedding the security controls, guardrails and tooling required to support the safe use of AI across Westpac. You will develop production‑grade code and reusable patterns to protect AI‑powered applications and agent‑based workflows, while working across application security and AI engineering to ensure solutions are secure by design. The role focuses on enabling the adoption of technologies such as Large Language Models (LLMs) and autonomous agents within a complex and highly regulated environment.

• →Design and implement security guardrails for AI and agent‑based applications, including controls for input/output filtering, prompt injection and agent permissions 
• →Develop production‑quality Python code for security controls, middleware and testing across LLM applications, RAG pipelines and multi‑agent systems 
• →Build reusable security patterns, libraries and reference implementations to support secure AI development 
• →Perform threat modelling and adversarial testing, covering risks such as data leakage, model evasion, prompt injection and agent misuse 
• →Embed automated security checks into AI/ML pipelines, including supply chain validation, dependency analysis and runtime monitoring 
• →Work with data, platform and engineering teams to integrate security into AI workflows and contribute to standards aligned with industry frameworks

• Proven experience implementing security guardrails in AI and agent‑based applications, including prompt safety, data protection and agent control mechanisms 
• Experience building and delivering reusable AI security libraries, patterns and reference implementations used by engineering teams 
• Hands‑on experience integrating automated security testing and validation into CI/CD and ML pipelines 
• Demonstrated experience identifying and reducing AI‑related security risks through prevention, detection and control implementation 
• Experience developing or contributing to AI security standards, guidance and enablement materials for engineering teams 
• Strong background in AI/LLM security, Python‑based security engineering, application security and threat modelling, with exposure to agentic frameworks, ML pipeline security and cloud platforms