- Lead incident response for product-level security events focused on AI-specific threats such as prompt injection, model abuse, agent hijacking, and AI-enabled data exfiltration
- Integrate incident response into Cortex, Snowflake Intelligence, and AI-powered developer experience pipelines from design through deployment
- Develop detection, containment, and remediation playbooks for LLM misuse, adversarial inputs, and AI-assisted attacks
- Address security and incident-response readiness technical debt across the AI product stack
- Represent incident response to cloud engineering, AI platform, corporate security, and customer-facing business units
- Secure AI-native codebases across multi-cloud environments, including inference services, RAG pipelines, vector stores, and agent orchestration layers
- Advise AI and security engineering teams on secure architecture for high-impact AI features
- Design and manage response capabilities across model-serving endpoints, Cortex Search indexes, and Snowpark ML pipelines
- Build data-, code-, and automation-driven tooling to accelerate product security incident detection and response
- Drive security outcomes for customers and enterprises using Snowflake AI workloads
Requirements
- 5+ years of experience in information security, primarily incident response, security engineering, or product/application security (preferred)
- Direct experience serving as incident commander for product-focused security incidents
- Experience leading or actively building an application or security engineering program
- Clear point of view on securing AI/ML systems
- Experience with threat modeling and security testing across AI attack surfaces, including prompt injection, indirect injection, model inversion, embedding extraction, and AI dependency supply-chain attacks
- Familiarity with data governance and security challenges involving LLMs, RAG architectures, and agentic systems
- Working knowledge of AWS, Azure, GCP, SaaS platforms, and AI platform threats
- SQL proficiency
- Experience building automation and tools with common programming languages, Python preferred
- Strong communication skills for translating security risk into actionable product-team guidance
- Bachelor's degree in Computer Science or a related field, or equivalent experience
- Bonus: experience securing AI/ML infrastructure, model serving, vector databases, embedding pipelines, API gateways, and LLM-integrated applications
- Bonus: experience building agentic incident response capabilities
- Bonus: understanding of attacker TTPs, adversarial ML, agent manipulation, and LLM jailbreaking
- Bonus: familiarity with CI/CD and secure AI release lifecycle patterns
- Preferred certifications include GCIA, GCIH, GCSA, GDAT, CISSP/GISP, or AWS, Azure, or GCP certifications
Core Competencies
Demonstrates expertise in incident response for AI-specific security threats, including prompt injection and model abuse, while integrating security practices into multi-cloud environments. Proficient in developing playbooks and tooling for AI security, with strong communication skills to guide product teams on security risks.
Highest-signal resume keywords
- Incident Response Leadership
- AI/ML Security Expertise
- Threat Modeling and Security Testing
- Automation and Tool Development
- Cloud Security Knowledge
ATS Optimization Keywords
Hard Skills
- Incident Response
- Security Engineering
- Threat Modeling
- SQL Proficiency
- Python Programming
- AI/ML Security
- Automation Development
- Security Testing
- Data Governance
- Incident Command
Soft Skills
- Strong Communication Skills
Certifications & Qualifications
- GCIA
- GCIH
- GCSA
- GDAT
- CISSP
- GISP
- AWS Certification
- Azure Certification
- GCP Certification
Industry Keywords
- AI Security
- Incident Response
- Model Serving
- Adversarial ML
- Agent Manipulation
- LLM Jailbreaking
- RAG Architectures
- AI Dependency Supply-Chain Attacks
- Embedding Pipelines
- Agentic Systems
Tools & Technologies
- AWS
- Azure
- GCP
- Cortex
- Snowflake
- SaaS Platforms
- CI/CD
- Vector Databases
- API Gateways
- LLM-integrated Applications