Job Description:
We are hiring a Senior Principal AI Security Architect to lead the technical security design for our AI initiatives—including LLM pipelines, retrieval systems, and agentic AI frameworks—across both our internal corporate and engineering ecosystem and our product. This role is deeply AI productivity and AI engineering-focused: you will architect secure agent behaviors, build guardrails around tool execution, and embed secure AI patterns directly into the software development lifecycle (SDLC).
What You’ll Do
Agentic AI & LLM Security
- Architect security controls for AI agents, including action authorization, tool-access policies, and sandboxed execution environments.
- Design agent orchestration patterns that prevent harmful or unintended actions, cross-tenant data access, and context bleed.
- Build verification layers for agent output, chain-of-thought protection, and safe action routing.
- Implement runtime guardrails around prompt injection, reasoning manipulation, self-escalation, and agent decision loops.
AI Integration Into SDLC
- Embed AI security into every stage of the SDLC, including secure model onboarding, threat modeling, automated AI security tests, and gated promotion of AI features.
- Build automated CI/CD checks for LLM features—prompt validation, policy enforcement, adversarial test suites, and red-team scenarios.
- Partner with engineering teams to define secure coding patterns for AI components, model interfaces, retrieval pipelines, and agent workflows.
- Integrate AI behavior monitoring into observability platforms to support detection engineering and post-deployment validation.
Core Architecture & Security Engineering
- Architect secure AI systems: inference services, RAG pipelines, embedding/indexing layers, and vector DBs.
- Build and secure model registries, orchestration systems, and service-to-model communication patterns.
- Conduct deep technical threat modeling for AI features, agent systems, and data flows.
- Partner on design reviews to ensure secure-by-default implementation of AI capabilities in the SaaS platform.
- Lead technical direction across engineering for secure AI adoption and scalable production deployment.
What You Bring
- 15+ years engineering or security architecture experience in cloud-native SaaS environments.
- Hands-on expertise with LLM integration, agentic AI workflows, vector databases, and RAG architectures.
- Strong engineering background in AWS/Azure/GCP/OCI, Kubernetes, microservices, and distributed systems.
- Deep understanding of adversarial ML, secure prompt design, agent risk mitigation, and model hardening techniques.
- Proficiency in Python, TypeScript, Go, or similar languages.
- Experience embedding security controls directly into developer workflows and CI/CD pipelines.
Nice to Have
- Experience designing agent permission frameworks, hierarchical agent structures, or multi-step decision pipelines.
- Experience with LangChain, LlamaIndex, or custom-built agent platforms.
- Background in AI red teaming or developing AI-specific testing frameworks.
Why Join Us
- Shape the future of secure AI engineering and agent-based automation within a leading SaaS platform.
- Build foundational architectures that enable powerful, safe, production-grade AI capabilities.
- Influence the strategic and technical direction of high-impact AI initiatives.