Senior AI Security Consultant (Outside IR35) BASED IN DUBAI

Full Description

THIS IS BASED IN DUBAI SO IF YOU ARE THERE CURRENTLY, IT WOULD BE PREFERABLE!

Costello & Reyes Group Ltd has been engaged by its client, an international consulting firm, to identify a Senior AI Security Consultant for 3 month initial period.

This engagement, if done well, WILL extend past this point for a further year.

We are seeking a highly skilled Senior AI Security Consultant to lead and matureour AI/ML security posture across enterprise-scale environments.

This role combines deep expertise in model security, adversarial resilience, and AI-specific threat detection with strong governance, compliance, and regulatory understanding.

You will act as the technical authority on securing AI/ML models, LLM integrations, and end-to-end AI pipelines, supporting teams across engineering, security, compliance, and product functions.

This is a strategic, hands-on position designed for an expert who can guide architecture, perform risk assessments, advise design decisions, and support incident response efforts relating to AI systems.

BASED in DUBAIKey Responsibilities1.

Model Security & Risk Assurance (Primary Need)Conduct adversarial threat assessments and evaluate model resilience against evasion, poisoning, and inference attacks.

Implement and advise on model explainability, fairness, and bias detection techniques to ensure ethical and transparent outcomes.

Develop and maintain model governance and model-versioning frameworks, ensuring traceability, auditability, and secure deployment practices.Review ML models for input/output manipulation risks, model drift, overfitting, and Embedded vulnerabilities.2.

Monitoring & AI Incident Response (Primary Need)Design and implement AI-aware monitoring capabilities, including anomaly detection for unusual model behaviour or performance deviation.

Develop AI-specific incident response playbooks, including escalation pathways for model compromise, hallucination events, data leakage, or prompt-based attacks.

Collaborate with SOC teams to define AI-focused alert thresholds, detection rules, and telemetry requirements.

Support post-incident analysis, including model forensics, root-cause assessment, and corrective control implementation.3.

External LLM/API Integration Security (Primary Need)Implement and review prompt security controls, including input sanitisation, output filtering, toxicity detection, and jailbreak-prevention measures.

Guide teams on secure LLM-to-LLM and LLM-to-service API design, including authentication, rate-limiting, encryption, and data isolation.

Assess risks relating to data leakage, context-bleed, Embedded secrets, and sensitive information exposure.

Evaluate and enforce safeguards for third-party AI provider usage, including contract, privacy, and technical control considerations.4.

AI Compliance, Governance & Ethics (Primary Need)Interpret and align programmes to relevant AI-related regulations, including EU AI Act, UK AI governance frameworks, ISO/IEC 42001, NIST AI RMF, and sector-specific obligations.

Integrate AI ethics principles (fairness, accountability, explainability, transparency) into development workflows and design standards.

Provide guidance on legal and compliance risks, including copyright, dataset provenance, model usage rights, and algorithmic accountability.

Support internal audit and external assessments covering AI governance maturity and control effectiveness.

Secondary Skill Areas (Awareness Required)Although covered by other teams, the successful candidate should have solid working knowledge and be able to support or advise when required in the following:Data Security & PrivacyUnderstanding of data life cycle protection, including encryption, masking, tokenisation, retention, and access controls.

Awareness of PII/PHI handling requirements, GDPR implications, and responsible dataset creation.

Experience supporting data access auditing, dataset provenance reviews, and consent-driven data management practices.

Pipeline & Infrastructure SecurityAwareness of ML pipeline and CI/CD security risks, including dependency vulnerabilities, supply-chain risks, and environment hardening.

Familiarity with container security, Kubernetes/ML Ops platforms, runtime protections, and secure model-hosting environments.

Understanding of secrets management, privileged-access control, and integration of security gates within ML pipelines.

Ability to collaborate with engineering on infrastructure threat modelling and secure deployment patterns.

Required Experience & QualificationsSignificant professional experience in AI/ML security, red-teaming ML models, or securing production AI environments.

Proven track record assessing and mitigating LLM/ML-specific threats (jailbreaks, adversarial perturbations, model extraction, data poisoning, inference attacks).

Strong understanding of AI architecture, transformer models, vector stores, embeddings, and model optimisation techniques.

Experience working with AI monitoring tools, drift-detection frameworks, and model-performance observability platforms.

Familiarity with relevant standards/frameworks such as NIST AI RMF, ISO/IEC 27001, ISO/IEC 42001, OWASP Top 10 for LLMs, MITRE ATLAS, and secure-ML design patterns.

Excellent communication skills with the ability to translate complex ML security concepts to technical and non-technical stakeholders.

THIS IS BASED IN DUBAI SO IF YOU ARE THERE CURRENTLY, THAT WOULD BE PREFERABLE!

Desirable QualificationsIndustry certifications such as:o Certified AI Security Specialist (CAISS)o GIAC Artificial Intelligence Security (GAISE)o GIAC Cloud Security (GCSA/GCPN)o Azure/AWS ML Specialityo CISSP, CCSP, CRISCo ISO 42001 Lead Implementer/AuditorExperience in regulated sectors (finance, healthcare, government, defence, energy).

Experience in AI red-teaming, prompt-engineering security, and model interpretability assessment tools.

Soft Skills & BehavioursStrategic thinker with the ability to shape and evolve an organisation's AI security roadmap.

Highly analytical, with exceptional problem-solving and forensic investigation skills.

Comfortable working autonomously in ambiguous environments, driving clarity and structure.

Strong stakeholder engagement and cross-functional collaboration skills.

Passionate about responsible AI, ethical governance, and emerging technology risk management.

This is an immediate need so should you be interested, please submit your profile. Costello & Reyes Group Ltd operate as a recruitment partner for its clientsTHIS IS BASED IN DUBAI SO IF YOU ARE THERE CURRENTLY, THAT WOULD BE PREFERABLE!