Lead AI Security Engineer

About Mastercard and Role Overview

Mastercard enables economies and empowers individuals across over 200 countries worldwide by delivering secure, straightforward digital payment solutions. Our innovative technologies and partnerships create unique products and services that unlock potential for individuals, businesses, and governments. We are seeking a Lead AI Security Engineer to join a strategic AI engineering team focused on foundational model development and deployment of AI use cases. This role specializes in hands-on security implementation for AI models and platforms, including responsible AI practices, vulnerability testing, and enforcing governance policies into production.

Key Responsibilities

• Translate and execute security recommendations from centralized security and AI governance teams into practical engineering solutions.
• Conduct vulnerability assessments of AI models, identifying risks such as membership inference, model inversion attacks, data leakage through embeddings or outputs, and adversarial input manipulation.
• Develop and sustain protective controls, including output filters, input validation, rate limiting, and access restrictions based on identified threats or governance requirements.
• Perform bias and fairness evaluations of AI models per guidance from AI ethics and governance teams.
• Produce technical documentation and reproducible evidence (test results, logs, benchmarks) to support internal governance and security audits.
• Monitor emergent model vulnerabilities and attack vectors, proactively testing platform defenses against novel threats.
• Serve as the primary technical liaison between the AI engineering team and company-wide security, privacy, and AI governance bodies, accurately communicating requirements and architectural details.
• Provide secure design recommendations to engineering teams during development phases, particularly on API, data pipeline, and model-serving components.
• Assist in incident response related to security or model vulnerabilities by reproducing, fixing, and verifying remediations.

Required Qualifications and Expertise

• Demonstrated experience in security testing of ML/AI models, including practical adversarial robustness assessments and red-teaming techniques like membership inference and model inversion.
• Strong engineering ability to implement technical security controls such as guardrails, filters, and access restrictions beyond advisory roles.
• Experience in conducting bias and fairness assessments and implementing mitigations for machine learning models.
• Understanding of security challenges specific to embeddings and foundation models, and awareness of cutting-edge research and attack methods in this area.
• Hands-on experience implementing access control and audit logging across hybrid or distributed computing environments (cloud and on-premises).
• Working knowledge of financial data regulations including PCI DSS and data protection laws sufficient to enforce correct security controls.
• Proficient software engineering skills to develop production-grade tooling and comprehensive test suites.
• Effective communication skills to collaborate with engineering peers and specialized advisory teams.
• Familiarity with cloud platforms, particularly AWS, and experience with data/AI platforms such as Databricks. Knowledge of Azure or GCP is an advantage.
• Aptitude for translating external governance and security guidance into precise, effective implementations suited to the platform’s architecture.

Corporate Security Responsibilities

• Adhere strictly to Mastercard’s security policies and practices.
• Maintain confidentiality and integrity of accessed information.
• Promptly report any suspected information security violations or breaches.
• Complete all mandatory periodic security training as per company guidelines.