Senior AI Cyber Security Engineer

Senior AI Cyber Security Engineer

Location : hybrid - 3 days onsite in Dunkirk, NY – try to find local candidates

Fulltime permanent role

What You’ll Be Doing

• Design and implement comprehensive security controls for AI/ML platforms, including model training environments, inference services, data pipelines, and feature stores

• Conduct threat modeling for AI systems to address model theft, data poisoning, prompt injection, model inversion, and abuse/misuse scenarios

• Build and maintain security tooling and automation to detect and prevent AI-specific attacks such as adversarial inputs, prompt injection chains, and anomalous usage patterns

• Collaborate with data scientists and ML engineers to integrate security throughout the AI development lifecycle, including secure coding practices, model validation, and red teaming for AI behaviors • Evaluate and harden integrations with third-party AI providers (LLM APIs, vector databases, orchestration frameworks, agents), ensuring proper authentication, authorization, data handling, logging

• Partner with cloud and platform teams to ensure AI infrastructure (Kubernetes, GPU clusters, model registries, CI/CD) follows security best practices and compliance requirements

• Define and implement monitoring for AI systems, including abuse detection, drift and anomaly alerts, model access patterns, and security-relevant telemetry

• Collaborate with incident response teams on investigations involving AI systems, analyzing logs, traffic, model behavior, and potential data or model compromise

• Provide technical guidance and mentorship to engineering teams on AI security concepts, threats, and secure design patterns

• Contribute to policies, standards, and guardrails for responsible and secure use of AI, including data classification rules for training inputs and prompts, allowed use cases, and evaluation requirements

What You’ll Need to Have

• 8+ years of hands-on security engineering experience (application, product, cloud, or infrastructure), including designing and implementing security controls in production environments

• Practical experience with AI/ML systems such as ML pipelines, LLM applications, vector search, or MLOps platforms, either as a security engineer or in close collaboration with ML teams

• Proven experience implementing authentication/authorization, secrets management, network segmentation, and secure CI/CD for services and APIs

• Strong understanding of modern cloud platforms (AWS, Azure, or GCP) and container/orchestration technologies (Docker, Kubernetes) as they apply to AI workloads

• Solid knowledge of common AI/ML threat scenarios including data poisoning, model theft/exfiltration, adversarial examples, model inversion, prompt injection, and jailbreaks

• Proficiency in Python (preferred) or other major programming languages used in security and ML ecosystems (Go, Java)

• Strong background in secure software development practices, code review, and security design reviews

• Ability to communicate complex technical risk and tradeoffs clearly to both technical and non-technical stakeholders