Principal AI Security Expert

Uney GmbH; Dubai; On-site; AED 400,000 - 600,000; Full time

Job summary

A tech firm specializing in AI security is seeking a Principal AI Security Expert to lead and manage the security of AI systems. This role involves designing security architectures, leading adversarial research, and mentoring team members in secure AI development. The ideal candidate will hold a PhD in a related field or a Master's with significant experience, possessing over 10 years in cybersecurity. Location is in Dubai, UAE. This hands-on role combines deep technical expertise with strategic oversight.

Qualifications

• PhD in a relevant field or Master's with extensive experience.
• 10+ years in cybersecurity or AI security roles.
• Proven leadership and a track record of remediating vulnerabilities.

Responsibilities

• Lead AI security architecture design and security audits.
• Conduct adversarial research and develop attack methodologies.
• Implement defenses against various attacks on AI systems.

Skills

Adversarial machine learning

Python programming

Cybersecurity

Deep learning frameworks (PyTorch, TensorFlow)

Leadership

Risk assessment

Education

PhD in Computer Science, Cybersecurity, or AI

Master’s degree with 10+ years professional experience

Job description

Lead and manage the security of artificial intelligence systems across the entire development lifecycle. Responsible for designing robust AI security architectures, leading adversarial research, implementing defenses against attacks on machine learning and Transformer models, and elevating team capabilities in secure AI development. This is a hands-on, technical leadership role combining deep expertise in AI security, practical implementation, and strategic oversight.

Key Responsibilities

AI Security Strategy & Architecture

• Design security architectures for AI systems, including threat modeling, vulnerability assessment, and risk mitigation frameworks.
• Lead security reviews and audits of AI models and systems throughout the development lifecycle.
• Collaborate with product, engineering, and data science teams to integrate security into AI development processes.
• Develop and maintain adversarial attack methodologies and tools for proactive security testing.

Adversarial Research & Red Teaming

• Lead red teaming exercises to identify and exploit vulnerabilities in AI models and systems.
• Conduct adversarial research to discover novel attack vectors against classical ML and Transformer models.
• Continuously evaluate AI system weaknesses and recommend mitigations to prevent exploitation.

Robust AI Development & Defense

• Research and implement techniques to enhance AI model robustness against adversarial attacks.
• Evaluate and implement defenses against common attacks, including evasion, poisoning, extraction, and membership inference.
• Design and implement guardrails to prevent harmful outputs and unsafe AI behaviors.
• Implement input validation, context injection prevention, and monitoring systems for anomalous model behavior.
• Develop defenses against prompt injection, context confusion, and semantic attacks.
• Guide technical direction for AI security initiatives and elevate team capabilities.
• Mentor team members on best practices in AI security, adversarial research, and defensive modeling.
• Collaborate cross-functionally to ensure AI systems are secure by design.

Qualifications

Education & Experience

• PhD in Computer Science, Cybersecurity, AI, or related field with research in AI security, adversarial ML, or related areas (or Master’s degree with 10+ years of professional experience).
• Minimum 10+ years of professional experience in cybersecurity, machine learning, or AI security.
• At least 8+ years in a senior or lead role with demonstrated technical leadership.
• Proven track record of identifying and remediating critical security vulnerabilities in AI systems.
• Published research or significant contributions in AI security.
• Expertise in adversarial machine learning and attack methodologies against classical ML models and Transformers.
• Hands-on experience with adversarial attack frameworks (FGSM, PGD, C&W, etc.).
• Proficiency in Python and deep learning frameworks (PyTorch, TensorFlow).
• Strong understanding of cryptography, secure computation, and privacy-preserving techniques.
• Knowledge of prompt injection, jailbreak, and context confusion attack vectors.
• Experience with security testing tools, vulnerability assessment methodologies, and risk mitigation.

Soft Skills

• Excellent leadership and team management capabilities.
• Ability to make high-stakes decisions under pressure.
• Strong communication and presentation skills for technical and non-technical audiences.
• Analytical thinking, problem-solving, and attention to detail.
• Customer- and risk-focused mindset, with the ability to balance security and usability.

You are applying for Principal AI Security Expert

Full name *

Email *

Phone *

How did you hear about UNEY? *

Profile reference

Upload your CV (PDF format) *

Add or drag & drop your CV here

Supported format: *.pdf, up to 10MB

By clicking "Submit" button, I agree that my information above will be used as described in UNEY's privacy policy