AI Security Engineer — Enterprise AI Guardrails Expert

Norton Blake; Greater London; On-site; GBP 150,000 - 180,000; Full time

Job summary

Norton Blake is seeking an experienced AI Security Engineer to join a leading investment bank on a 6-month contract. You will design, assess and deploy secure AI and GenAI solutions across the organisation, advising on secure architecture, third-party risk and governance.

The role involves designing security guardrails for prompt injection, data exfiltration and model poisoning, and collaborating with Engineering, Legal, Risk and Compliance to embed secure AI practices across delivery teams.

Qualifications

• Experience assessing AI/ML systems from a security perspective.
• Knowledge of AI security frameworks (OWASP Top 10 for LLM Apps, NIST AI RMF, MITRE ATLAS).
• Experience auditing external AI vendors and APIs.
• Ability to design security guardrails for AI solutions including prompt injection and data leakage.

Responsibilities

• Analyse AI/ML systems and architectures from a security perspective.
• Conduct security assessments, threat modelling and risk reviews for internal AI solutions and external vendors.
• Design security guardrails to protect against prompts, data exfiltration and model poisoning.
• Assess third-party AI APIs, platforms and suppliers for proper security and data protection controls.
• Define secure patterns for AI API integrations, authentication, identity management and data-in-transit protection.
• Translate AI security risks into practical recommendations for stakeholders.
• Collaborate with Engineering, Architecture, Legal, Risk and Compliance to embed secure AI practices.
• Support alignment with AI security frameworks and governance standards.

Skills

AI security

Security architecture

Threat modelling

API security

OAuth

IAM

Tools

NeMo Guardrails

Llama Guard

Vertex AI

Azure OpenAI

Job description

Norton Blake is seeking an experienced AI Security Engineer to join a leading investment bank on a 6-month contract. You will design, assess and deploy secure AI and GenAI solutions across the organisation, advising on secure architecture, third-party risk and governance.

The role involves designing security guardrails for prompt injection, data exfiltration and model poisoning, and collaborating with Engineering, Legal, Risk and Compliance to embed secure AI practices across delivery teams.