AI Security Lead: GenAI Data Protection & Gateway

Citizens Bank; Pittsburgh (Allegheny County); Hybrid; USD 130,000 - 195,000; Full time

Job summary

Citizens Bank is seeking a Principal Security Engineer to lead the design, deployment, and enterprise rollout of its AI Security and GenAI data protection capabilities. This role owns the AI Gateway platform and drives policy engineering, remediation, and governance across sanctioned and unsanctioned AI usage.

You will collaborate with Cybersecurity, Data Protection, Risk, Legal and business teams to reduce data loss risk while enabling responsible AI adoption, with a hybrid work model and

Qualifications

• Bachelor's degree in Computer Science, Information Security, Information Technology, or related field.
• 7+ years of progressive experience in cybersecurity, information security engineering, or related technical disciplines.
• 2+ years of hands on experience with Data Protection, SSE, SASE, CASB, DLP, AI Security, GenAI Security, or similar inline security technologies.
• Direct AI Gateway experience strongly preferred.
• Deep understanding of GenAI data flows and security risks across prompts, uploads, browser traffic, embedded copilots, APIs, AI agents, and third party integrations.
• Experience implementing enterprise security controls capable of monitoring, coaching, redacting, quarantining, or blocking risky data movement in real time.

Responsibilities

• Own the end to end AI Gateway capability across evaluation, proof of concept, selection, deployment, configuration, policy tuning, operational support, and continuous improvement.
• Develop and execute a phased rollout from monitor mode to warn and coach to targeted blocking, using observed AI traffic and risk metrics to establish baselines and control implementation.
• Engineer and tune controls for GenAI prompts, file uploads, browser traffic, embedded copilots, and sanctioned and unsanctioned AI applications.
• Implement inline remediation controls including redaction, quarantine, user coaching, warning, and blocking.
• Establish instance aware controls that distinguish corporate from personal accounts and reduce data leakage risk.
• Assess AI and agentic integration risks involving AI agents, Model Context Protocol (MCP), non human identities, third party integrations, and emerging access paths.
• Integrate AI Gateway telemetry with SIEM, SOAR, ITSM, identity platforms, data protection solutions, and evidence repositories.
• Define and operationalize governance standards, policy lifecycle processes, exception handling procedures, escalation paths, and ownership models.
• Measure and report program effectiveness through coverage, policy accuracy, false positive and false negative trends, remediation outcomes, audit evidence quality, and user experience metrics.
• Partner with business leaders and AI platform owners to introduce coaching and enforcement controls with minimal disruption.
• Stay current on AI security threats, GenAI data loss trends, agentic AI developments, MCP capabilities, and evolving security technologies.
• Contribute to standards, procedures, training materials, and enterprise communications related to secure AI adoption.
• Support risk, audit, and regulatory readiness through maintenance of defensible security evidence, policy decisions, exceptions, and enforcement outcomes.

Skills

AI Security

GenAI Security

SIEM/SOAR

Policy governance

Risk assessment

Cross-functional leadership

Education

Bachelor's degree in Computer Science or related field

Tools

SIEM

SOAR

ITSM

OAuth/OIDC

DLP / Data Loss Prevention

Job description

Citizens Bank is seeking a Principal Security Engineer to lead the design, deployment, and enterprise rollout of its AI Security and GenAI data protection capabilities. This role owns the AI Gateway platform and drives policy engineering, remediation, and governance across sanctioned and unsanctioned AI usage.

You will collaborate with Cybersecurity, Data Protection, Risk, Legal and business teams to reduce data loss risk while enabling responsible AI adoption, with a hybrid work model and