Global Red Team AI Engineer, AVP

Mitsubishi UFJ Financial Group•Jersey City, NJ

7d•Hybrid

About The Position

The Assistant Vice President, Red Team & AI Security supports the planning, execution, and reporting of adversary simulation, control validation, and AI security assessment activities across the enterprise. This role is hands-on and operator-focused, combining offensive security tradecraft with AI analysis responsibilities to help identify exploitable weaknesses, validate defensive controls, and translate findings into practical remediation guidance. The selected colleague will work at an MUFG office or client sites four days per week and work remotely one day. A member of our recruitment team will provide more details.

Requirements

4+ years of experience in offensive security, red team operations, penetration testing, application security, cloud security, cyber defense, or related security assessment roles.
Hands-on experience supporting security assessments, adversary emulation, control validation, vulnerability analysis, or threat-informed testing.
Working knowledge of common attack techniques across Windows, Linux, Active Directory, web applications, APIs, cloud platforms, and identity systems.
Ability to write or modify scripts and tools using languages such as Python, PowerShell, Bash, JavaScript, or similar.
Familiarity with offensive security tools, vulnerability validation, command-and-control concepts, detection engineering concepts, and MITRE ATT&CK mapping.
Understanding of AI security risks such as prompt injection, excessive agency, insecure tool use, sensitive data exposure, insecure plugins, model abuse, and AI-enabled social engineering.

Nice To Haves

Exposure to AI security, LLM testing, AI governance, data security, machine learning workflows, or secure use of AI-assisted tools is preferred.
Bachelor’s degree in Cybersecurity, Computer Science, Information Technology, Engineering, Data Science, or equivalent experience.
Security certifications such as OSCP, GPEN, GWAPT, GCIH, GCIA, GSEC, CISSP, or similar are preferred but not required.
Familiarity with regulatory, risk, or control frameworks such as NIST, FFIEC, GLBA, SOX, PCI, or internal control standards is beneficial.

Responsibilities

Support the planning and execution of Red Team, Purple Team, adversary simulation, and control validation engagements.
Assist with target research, scope development, rules of engagement, infrastructure preparation, and operational documentation.
Coordinate with defensive, engineering, risk, and business stakeholders to ensure engagements are well-structured and findings are actionable.
Perform hands-on offensive security testing across network, endpoint, identity, application, cloud, and selected physical attack surfaces.
Execute techniques such as reconnaissance, initial access validation, privilege escalation, lateral movement, persistence simulation, and control evasion in approved environments.
Develop, modify, and safely use scripts, payloads, tooling, and tradecraft to support realistic adversary emulation.
Support adversarial testing of AI and LLM systems, including prompt injection, data leakage, tool misuse, model behavior abuse, and guardrail bypass scenarios.
Analyze AI-enabled applications, copilots, automation workflows, agents, plugins, APIs, and data integrations for security and abuse risks.
Document AI security observations in clear business and technical terms, including risk impact, exploitability, control gaps, and recommended remediation.
Use AI-assisted analysis responsibly to accelerate research, data review, reporting, detection mapping, and test-case development.
Track emerging AI threats, attacker use of AI, and relevant industry guidance to inform Red Team scenarios and validation priorities.
Drive research into real‑world threat actor TTPs and emerging AI‑enabled attack techniques.
Sponsor development of red team tools, scripts, and repeatable methodologies to increase realism, scale, and efficiency.
Establish standards for tooling governance, safe use, and audit readiness.
Prepare engagement notes, evidence packages, attack narratives, findings, remediation guidance, and executive-ready summaries.
Map observations to relevant frameworks, controls, detections, and business risk themes where appropriate.
Partner with blue team, cyber defense, engineering, governance, and application teams to validate remediation and improve control effectiveness.