AI Security Research Engineer II

AI Security Engineer II – (United States) Meet the Team

The mission of the Security & Trust Organization (S&TO) is to secure products and services, protect Cisco and its employees, and cultivate trust with our customers, partners, and regulators.

Thrive on our dynamic and collaborative team where you'll work closely with a diverse network of internal and external stakeholders to drive essential security initiatives. You'll have the opportunity to collaborate with teams across Cisco, including IT, Product, Customer Experience (CX), and nearly every business unit, to deliver critical security solutions.

As an AI Security Engineer I, you will help advance the secure adoption of artificial intelligence across Cisco products, services, and internal operations. Working at the intersection of AI, security, and software engineering, you will partner with engineering, product, and security teams to identify and mitigate AI-related risks, support threat modeling and security assessments, develop automated security controls and validation mechanisms, and contribute to secure development practices for AI-enabled systems.

You will also research emerging AI attack techniques and defensive strategies, help build reusable security standards and enablement resources, and support the responsible deployment of AI technologies across the organization.

• Completion within the past 3 years, or current enrollment with expected completion within 12 months, of a Bachelors + 2 years of relevant experience or Masters + 0 years of relevant experience. Relevant fields include:
  Cybersecurity, Computer Science, Software Engineering, Artificial Intelligence or Machine Learning.
• Foundation in security engineering (application/product security, secure SDLC, threat modeling) and/or AI/ML engineering – strong in one area with working exposure to the other with demonstrated adversarial thinking.
• Proficiency in Python and experience developing scripts, automation, applications, or security tooling.
• Demonstrated analytical and problem‑solving skills with an interest in adversarial thinking, emerging technologies, and securing AI‑enabled systems.
• Must be a U.S. citizen with authorization to work in the U.S. without current or future need for visa sponsorship.

• Hands‑on experience securing or evaluating Large Language Models (LLMs), including prompt injection defenses, guardrails, insecure output handling, RAG, or agentic AI systems.
• Familiarity with AI security frameworks and standards such as OWASP LLM Top 10, OWASP Agentic Top 10, MITRE ATLAS, NIST AI RMF, ISO/IEC 42001, or the EU AI Act.
• Experience with AI security testing, red teaming, evaluation frameworks, Cisco AI Defense, NVIDIA Garak, or similar AI security tools.
• Experience with CI/CD pipelines, MCP, agent‑to‑agent integration patterns, or AI‑assisted development tools.
• Demonstrated passion for security through applied projects, security tooling contributions, CTF participation, published research, or open‑source contributions.