Associate Director, AI Security Frontier Engineering

KPMG LLP; Montvale (NJ); On-site; USD 150,000 - 210,000; Full time

Job summary

KPMG LLP in Montvale, NJ is seeking an Associate Director for AI Security Frontier Engineering within Enterprise Security Services. You will own security architecture for frontier AI platforms and drive secure deployment patterns across cloud environments.

Lead end-to-end security engineering initiatives, establish reusable patterns, enhance telemetry, and mentor peers. Strong communication with leadership and engineering teams is required, with authorized work eligibility in the U.S.

Qualifications

• 8+ years of progressive security engineering or cloud security experience.
• Hands-on experience securing AI/ML/LLM systems and performing adversarial testing.
• Familiarity with NIST AI RMF, MITRE ATLAS, or OWASP Top 10 for LLMs.
• Experience designing controls across AWS, Azure, or GCP; logging/observability.

Responsibilities

• Serve as SME for frontier-AI platform security architecture and secure deployment patterns.
• Design and implement automated controls and guardrails per NIST CSF and OWASP Top 10 for LLMs.
• Lead end-to-end security engineering initiatives from risk assessment to release criteria.
• Create reusable patterns and frameworks to accelerate cross-platform security workstreams.
• Develop security telemetry, logging standards, and threat-informed detection integrated with SIEM/SOC.
• Coordinate with CSIRT and leadership to ensure durable security outputs.

Skills

Security engineering

Cloud security

AI/ML security

Threat detection

SIEM/logging

Adversarial testing

Communication

Education

Bachelor's degree in CS/Cybersecurity

High school diploma or GED required

Tools

AWS

Azure

GCP

NIST AI RMF

MITRE ATLAS

OWASP Top 10 for LLMs

Job description

KPMG is currently seeking an Associate Director, AI Security Frontier Engineering to join our Enterprise Security Services organization.

Responsibilities:

• Serve as the primary technical subject matter expert for a designated frontier-AI platform (for example, Anthropic, OpenAI, Microsoft, or Google), owning its security architecture and secure deployment patterns; sequence work against firm priorities and adapt patterns as each platform matures
• Design and implement automated controls and guardrails across the NIST CSF functions and OWASP Top 10 frameworks, such as mitigating prompt injection, preventing data leakage, and establishing secure model- and tool-interaction patterns
• Lead complex security engineering initiatives end-to-end with minimal oversight, from risk and gap assessment through control build-out to release criteria; deliver high-quality, defensible capabilities on schedule while operating comfortably in ambiguity
• Create reusable, cross-platform architecture patterns, control frameworks, and engineering standards so controls built for one AI ecosystem accelerate the next and reduce duplicated effort across concurrent workstreams
• Lead the development and maturation of security telemetry, monitoring, and detection engineering capabilities by establishing logging standards, onboarding security-relevant data sources, and delivering threat-informed detection content integrated with SIEM and SOC operations; partner with Adversarial Testing teams to validate detection effectiveness, identify visibility gaps, and strengthen defensive controls
• Establish non-negotiable MVP security and release criteria and hand durable outputs into operations by coordinating detection logic and playbook development with CSIRT and adjacent teams; serve as a technical advisor and mentor, communicating risks and solutions clearly to leadership and engineering peers
• Act with integrity, professionalism, and personal responsibility to uphold KPMG’s respectful and courteous work environment

Qualifications:

• Minimum eight years of recent progressive experience in security engineering or cloud security, including direct, hands-on experience securing enterprise frontier-AI platforms (such as Anthropic Claude, OpenAI ChatGPT Enterprise/Codex, Microsoft M365 Copilot, or Google Vertex AI/Gemini)
• Bachelor's degree from an accredited college or university in computer science, Cybersecurity, or a related technical field is preferred; minimum of a high school diploma or GED is required; relevant industry certifications such as CISSP, CCSP, or related cloud credentials are highly preferred
• Hands-on expertise securing AI/ML or LLM systems and executing adversarial testing (targeting prompt injection, jailbreaks, data leakage, and tool, plugin, or connector abuse), combined with familiarity with frameworks like NIST AI RMF, MITRE ATLAS, or OWASP Top 10 for LLMs
• Demonstrated ability to design and implement security controls across AWS, Azure, or GCP, with practical experience in agentic runtime, prompt-inspection controls, sandboxing, blast-radius containment, and modern logging or observability tooling
• Proven track record of independently leading technical initiatives end-to-end and resolving ambiguous, complex problems with minimal oversight, ideally operating in a fast-moving "tiger team" or professional services/highly regulated environment; strong working knowledge of detection engineering and security telemetry (SIEM and logging pipelines), alongside essential enterprise controls including DLP, identity/SSO, and endpoint management
• Excellent communication and collaboration skills to translate complex technical risks for both engineering teams and non-technical senior stakeholders
• Applicants must be authorized to work in the U.S. without the need for employment-based visa sponsorship now or in the future; KPMG LLP will not sponsor applicants for U.S. work visa status for this opportunity (no sponsorship is available for H-1B, L-1, TN, O-1, E-3, H-1B1, F-1, J-1, OPT, CPT or any other employment-based visa)