Lead AI Security Engineer

Start free →

Or apply yourself on the official page →

About the role

Our Purpose Mastercard powers economies and empowers people in 200+ countries and territories worldwide. Together with our customers, we’re helping build a sustainable economy where everyone can prosper. We support a wide range of digital payments choices, making transactions secure, simple, smart and accessible. Our technology and innovation, partnerships and networks combine to deliver a unique set of products and services that help people, businesses and governments realize their greatest potential. Title and Summary Lead AI Security Engineer Overview Mastercard is seeking a Lead AI Security Engineer to support a strategic AI engineering team focused on foundation model development and AI use case delivery. This role is responsible for the hands-on implementation of model security, AI product platform security and responsible-AI practices. Testing models for vulnerabilities, building the guardrails, access controls, API security, secure data usage and mitigations recommended by Mastercard's central security and AI governance teams, and ensuring those recommendations are actually built into the platform, not just documented. This is an execution-focused engineering role: Mastercard's dedicated security and governance teams already own policy, advisory guidance, and formal review. This role is the team's technical owner for turning that guidance into working, tested implementation. Role In this role, you will be responsible for implementing and maintaining the technical controls, tests, and mitigations that keep the platform's models secure and its AI use responsible in practice. Along with ensuring our developers, data engineers and AI engineers are incorporating security standards into their builds. Key responsibilities: Implement the technical recommendations produced by central security and AI governance review teams. Translating advisory guidance into concrete engineering work, and tracking it through to completion. Test models for security vulnerabilities directly. Running practical assessments for risks such as membership inference, model inversion, data leakage through embeddings or outputs, and adversarial input manipulation. Build and maintain guardrails and mitigations identified through vulnerability testing or governance review. For example, output filtering, input validation, rate limiting on sensitive queries, or access restrictions tied to specific risk findings. Implement bias and fairness testing for models and use cases, working from criteria and guidance set by central AI ethics/governance teams Build technical evidence and test results to support governance and security reviews. Providing concrete, reproducible proof (test results, logs, benchmark outputs) that central governance teams need. Stay current on emerging model vulnerability research and attack techniques (e.g., new inversion or extraction methods relevant to embeddings and transformer-based models) and proactively test the platform against them. Partner closely with central security, privacy, and AI governance teams as the team's primary technical point of contact. Understanding their requirements accurately and representing the platform's architecture to them clearly. Advise engineering peers on secure design choices during development, providing practical, specific input on API, data pipeline, and model-serving designs before they reach formal review. Support incident response for security or model-vulnerability findings, including reproducing issues, implementing fixes, and verifying remediation. All About You Required skills and experience: Hands-on experience testing ML/AI models for security vulnerabilities. Adversarial robustness testing, membership inference, model inversion, or similar practical red-teaming of models, not just fa

View the official posting →