Sr. Engineer, AI Cyber Security

Biogen is a biotechnology company focused on delivering life-changing medicines. The Senior AI Cyber Security Engineer will secure the organization’s AI, machine learning, and generative AI technologies by designing and operating AI-specific security controls, detection capabilities, and governance processes. The role also performs risk assessments, adversarial testing, security automation, and cross-functional collaboration with AI/ML and cybersecurity teams.

Responsibilities:

• Design, build, and operate detection capabilities for AI-specific threats including prompt injection, model extraction, training data poisoning, and adversarial input attacks
• Develop and maintain AI security monitoring pipelines that correlate signals across LLM interactions, API gateways, and model inference endpoints
• Engineer automated response playbooks for AI security incidents — including model quarantine, token revocation, and session termination
• Continuously research emerging AI attack vectors and translate findings into detection signatures and prevention controls
• Operate and tune CrowdStrike AI Runtime Defense (AIRD) and Microsoft Purview AI governance controls across the environment
• Implement security controls across AI development and deployment platforms (AWS SageMaker, Azure OpenAI, Databricks, internal ML pipelines)
• Secure model training environments — including data access controls, compute isolation, artifact signing, and pipeline integrity verification
• Engineer guardrails for generative AI usage — content filtering, DLP integration, output validation, and session-level access controls
• Design and enforce AI model governance controls including model registry security, version control integrity, and deployment approval gates
• Perform security architecture reviews of new AI services and integrations before production deployment
• Conduct technical risk assessments of AI systems using MITRE ATLAS, OWASP AI Top 10, and NIST AI RMF frameworks
• Perform adversarial testing (red teaming) of AI models and AI-integrated applications to identify vulnerabilities before production deployment
• Develop and maintain AI security standards, reference architectures, and secure development guidelines for AI/ML teams
• Evaluate third-party AI services and vendor AI integrations for security posture, data handling practices, and supply chain integrity
• Build and maintain AI-powered security tools that augment the cybersecurity team's detection and response capabilities
• Develop integrations between security platforms (CrowdStrike, Microsoft Defender, Okta, Zscaler) and AI/ML systems for intelligent threat correlation
• Engineer automated security workflows using LLMs and agentic AI for threat hunting, alert triage, and incident investigation acceleration
• Contribute to the security team's internal AI capabilities — including MCP server development, agent frameworks, and security data pipelines
• Build and maintain security scoring engines and risk quantification models that drive prioritization decisions

Requirements:

• 5+ years of combined experience in cybersecurity engineering, software engineering, and/or AI/ML systems
• Demonstrated expertise in at least three of the following domains:
• AI/ML Security: LLM security, prompt injection mitigation, model robustness testing, adversarial ML, AI supply chain integrity
• Cloud Security Engineering: AWS, Azure, or GCP security architecture and controls implementation at scale
• Application Security: Secure SDLC, API security, code review, penetration testing, threat modeling
• Security Operations: Detection engineering, SIEM/SOAR development, incident response, threat hunting
• Hands-on proficiency with Python and demonstrable experience building production security tooling and automation
• Experience with AI/ML frameworks (PyTorch, TensorFlow, Hugging Face, LangChain) and AI cloud services (SageMaker, Azure OpenAI, Bedrock, Databricks)
• Working knowledge of AI security frameworks: MITRE ATLAS, OWASP AI Top 10, NIST AI RMF
• Experience with enterprise security platforms (CrowdStrike, Microsoft Defender, Okta, Zscaler, CyberArk, or equivalent)
• Strong understanding of LLM architectures, retrieval-augmented generation (RAG), fine-tuning security implications, and agentic AI patterns
• Bachelor's Degree in Computer Science, Cybersecurity, Data Science, or related technical field required