REGISTRIEREN

Dieses Stellenangebot ist nicht mehr verfügbar

XX

AI Security Architect - Erlanger, KY

AgHires
  • US
    Erlanger, Kentucky, United States
  • US
    Erlanger, Kentucky, United States

Über

AI Security Architect - Erlanger, KY Position Summary ADM’s Global Information & Cyber Security (GICS), Security Architecture & Engineering team is seeking an AI Security Architect with an Engineering/Analyst background. This role safeguards enterprise AI systems by applying industry guidance on AI risk standards and principles. The position ensures secure, resilient, and compliant AI adoption across cloud and enterprise environments, focusing on confidentiality, integrity, availability, safety, and ethical use of AI/ML systems. The role proactively identifies and mitigates risks from adversarial machine learning, data poisoning, model leakage, and unauthorized access, while collaborating cross‑functionally to build secure and trustworthy AI systems.
Key Responsibilities
Consult, recommend and implement practices aligned with joint internal and external guidance, including understanding AI risks, securing the AI lifecycle, ensuring resilience, and establishing accountability.
Threat Modeling & Adversarial Testing: Conduct threat modeling for AI/ML models and pipelines, lead adversarial testing, red teaming, and stress testing on AI models.
Review current internal capabilities, processes, tooling and provide strategic and tactical recommendations to meet requirements to secure, defend, and thwart AI capabilities.
Threat Detection & Response: Monitor AI systems for adversarial ML attacks, prompt injection, and misuse.
Documentation & Best Practices: Develop and maintain documentation for AI security best practices.
Cross‑team Collaboration: Partner with AI engineers, architects, compliance officers, technologists, and stakeholders to embed controls and guide secure AI development.
Continuous Improvement: Stay up to date with advancements in AI and automation technologies to continuously improve security engineering.
AI Risk Management: Apply AI risk management standards to assess and mitigate risks in AI pipelines.
Required Skills & Experience
Strong knowledge of CISA Secure AI principles and ISO/IEC 23894.
Hands‑on experience with Microsoft Purview, Defender for Cloud, Entra ID, and Sentinel.
Understanding of AI/ML fundamentals (model training, inference, adversarial ML, secure data pipelines).
Experience with ML platforms (TensorFlow, PyTorch, Scikit‑learn) and MLOps tools (MLflow, Kubeflow).
Familiarity with adversarial ML concepts and tools (Pyrit, IBM Adversarial Robustness Toolbox, CleverHans).
Proficiency in scripting or programming languages (Python, Bash, .Net).
Knowledge of security tools and frameworks (STRIDE, MITRE ATLAS, vulnerability scanners, SIEM).
Expertise in securing workloads in Azure; AWS/GCP experience is a plus.
Ability to assess and mitigate AI‑specific risks (bias, poisoning, data leakage).
Familiarity with regulatory frameworks (GDPR, HIPAA, FedRAMP, CCPA).
Strong analytical, communication, and documentation skills.
Ability to explain complex AI security concepts to technical and non‑technical audiences.
Collaborative mindset with experience working across multidisciplinary teams.
Preferred Qualifications
5+ years in cybersecurity, with at least 2 years focused on AI/ML or cloud security.
Certifications: Azure Solutions Architect, GIAC Machine Learning Security Essentials (GMLE), CISSP, CCSP, or equivalent AI/ML security credentials.
Project management experience.
Experience with lifecycle and licensing within cloud environments.
Current holder of security certifications.
ISO/IEC 23894
CISA Secure AI Principles
NIST IR 8596 guidance and Cybersecurity Framework 2.0
Leadership Traits
Ownership mindset.
Commitment to helping others thrive.
Continuous learning.
Fostering diversity, equity, and inclusion.
ADM requires the successful completion of a background check.
#J-18808-Ljbffr
  • Erlanger, Kentucky, United States

Sprachkenntnisse

  • English
Hinweis für Nutzer

Dieses Stellenangebot wurde von einem unserer Partner veröffentlicht. Sie können das Originalangebot einsehen hier.