Ripple Senior Staff Security Engineer at Ripple focusing on AI Security to secure AI systems and leverage AI for security improvements. Lead AI security strategy, design controls for LLM and agentic AI, and shape external AI security posture.
Responsibilities
Drive the AI Security technical strategy and roadmap, defining how Ripple secures its AI systems, governs agentic workflows, and embeds security controls into the AI development lifecycle from day one.
Design and implement security controls for LLM-integrated and agentic AI systems, including sandboxing, identity and permission scoping, runtime monitoring, and containment of autonomous agent actions that exceed authorized scope.
Own AI security across the Controlled Agentic SDLC, establishing security guardrails, AI provenance standards, dual-review requirements, and audit trail controls for AI-assisted development across Ripple Engineering.
Lead the security review and risk assessment of all AI integrations entering production, including LLM APIs, SaaS copilots, AI code editors, agentic workflows, third-party MCP servers, and vendor-embedded AI.
Build and scale Ripple's Shadow AI detection capability, surfacing unsanctioned AI usage, driving adoption of the AI acceptable use policy, and ensuring all AI workflows operate within Ripple's auditable perimeter.
Serve as Ripple's go-to technical resource on agentic AI risks, including MCP server security, tool poisoning, prompt injection at the orchestration layer, and excessive agency in multi-agent systems, translating emerging threats into concrete mitigations with Engineering and Product.
Shape Ripple's external AI security posture, contributing to industry frameworks, engaging regulators, and publishing research that establishes Ripple as a credible voice in responsible AI security.
WHAT YOU'LL BRING:
10+ years of Security Engineering experience with demonstrated depth in at least two domains, such as Product Security, Cloud Security, or Security Operations, and meaningful hands-on exposure to AI or ML security in practice.
Solid understanding of AI and LLM security concepts, including prompt injection, jailbreaks, data poisoning, model extraction, RAG manipulation, and agentic risks such as tool poisoning, excessive agency, and MCP server vulnerabilities.
Experience securing agentic AI systems, including sandboxing, permission scoping, human-in-the-loop design, or runtime monitoring for autonomous workflows.
Fluency with core Security Engineering domains including cloud security on AWS, GCP, or Azure, CI/CD pipeline security, container and Kubernetes security, IAM, and API security, with the ability to reason about how these apply in AI-specific contexts.
