> Markdown version of [/jobs/ext/4629418-principal-ai-security-engineer](https://www.wearedevelopers.com/jobs/ext/4629418-principal-ai-security-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Principal AI Security Engineer - **Company:** Zip's Service - **Location:** United States (Remote available) - **Salary:** $203,000.0 - $254,000.0 - **Contract:** Permanent contract - **Skills:** LangGraph Framework, Application Programming Interfaces (APIs), Artificial Intelligence, Architectural Patterns, Automation of Tests, Microsoft Azure, Cloud Computing, Cloud Engineering, Cyber Security, Databases, Information Leak Prevention, Identity and Access Management, Python (Programming Language), Key Management, Machine Learning, Open Source Technology, Open Web Application Security, Regression Testing, OpenAI, Software Engineering, Google Cloud, Enterprise Software Applications, LangChain, Retrieval-Augmented Generation, Large Language Models, Multi-Agent Systems, Google Vertex AI, AI Coding Agents, Agentic-AI, Kubernetes, Azure AI, Terraform, Semantic Kernel, Security Orchestration, Automation & Response, Vulnerability Analysis, Programming Languages - **Published:** November 7, 2026 - **Apply:** https://rebrand.ly/check-zip-out ## About the Role * Deep security engineering experience plus demonstrable hands-on work with LLM/agent systems (production, open source, internal platforms, or published research). * Intimate familiarity with OWASP Top 10 for LLM Applications, MITRE ATLAS, and NIST AI RMF. * Strong understanding of LLM architectures, agentic systems, RAG, and modern AI development frameworks. * Experience building AI security evaluations, red-teaming capabilities, model security gateways, as well as securing AI systems against prompt injection, data leakage, tool abuse, and unauthorized access. * Strong software engineering skills, including proficiency in Python or another modern programming language. * Experience with cloud-native architectures, APIs, containers, IAM, and modern software delivery practices. * Ability to design and implement security controls, automation, and reusable engineering components. * Experience threat modeling complex systems and translating security requirements into practical solutions. * Strong technical leadership skills with the ability to influence architecture and engineering decisions across teams. * A builder's mindset focused on automation, scalability, and secure-by-default engineering., * Experience with LangChain, LangGraph, Semantic Kernel, OpenAI APIs, Azure AI, Google Vertex AI, or similar technologies. * Experience securing MCP servers, autonomous coding agents, or agentic workflows. * Familiarity with OWASP Top 10 for LLM Applications, MITRE ATLAS, and NIST AI RMF. * Experience with Microsoft Azure, Google Cloud Platform (GCP), Terraform, Kubernetes, and CI/CD security. * Experience in financial services, payments, or other regulated environments. ## Description * Lead the technical direction for securing AI applications, autonomous agents, and AI-enabled engineering workflows at Zip. * Architect and build scalable security controls for LLM applications, RAG systems, and agentic AI platforms. * Develop AI security testing, automation, and secure-by-default engineering practices to enable safe innovation at scale. * Remote-first opportunity for US-based employees, with the option to work in person from our Manhattan office. Start your adventure with Zip Join Zip's Security team and help shape how we securely build and operate AI-powered systems. AI is transforming how we develop software, operate our platforms, and deliver products. As our Principal AI Security Engineer, you'll lead the technical direction for securing AI applications, autonomous agents, and AI-enabled engineering workflows. This is a hands-on engineering role for someone who has built and secured AI solutions in production. You'll work directly with engineering, platform, and security teams to design and implement scalable security controls that enable innovation without introducing unnecessary risk. We're looking for a builder who can establish architectural direction, write code, and turn complex AI security challenges into practical engineering solutions., * Define and implement security architecture for LLM applications, RAG, autonomous agents, and AI-powered products. * Engineer protections against prompt injection, data exfiltration, insecure tool execution, excessive agency, and other AI-specific threats. * Establish secure patterns for model access, NHI access, authorization, secrets management, and sensitive data protection. * Design security boundaries for AI agents interacting with enterprise systems, APIs, databases, and cloud infrastructure. Build security into AI-first engineering * Establish secure-by-default patterns that allow developers to adopt AI without bypassing security controls. * Develop secure harnesses and guardrails for AI coding assistants, autonomous development agents, and AI-generated software. * Secure MCP servers, agent tool integrations, and AI-enabled development workflows. * Build reusable security components, APIs, and automation that integrate directly into engineering platforms and CI/CD pipelines. Engineer AI security testing and automation * Build automated security testing and evaluation capabilities for AI applications and agents. * Develop approaches for AI red teaming, prompt injection testing, and security regression testing. * Implement monitoring and detection for suspicious AI behavior, unauthorized tool execution, and sensitive data exposure. * Post incidence response - agent misalignment (exceeds authority, unauthorized action). * Partnering with the Agentic Platform team on controls to prevent recurrence. * Leverage AI to automate security engineering activities, including threat modeling, vulnerability analysis, and remediation. Lead AI security architecture * Establish Zip's technical direction and engineering standards for AI security. * Partner with engineering teams to take secure AI solutions from architecture through production. * Design risk-tiering logic aligned with Zip's 5-level AI automation classification. * Co-own the security evaluation framework for new AI models and third-party tools. Evaluate emerging AI technologies and security capabilities, making pragmatic build-versus-buy decisions. * Mentor engineers and influence technical decisions across security, platform engineering, and software development., * Engineering teams can securely build and deploy AI solutions using reusable security controls and established architectural patterns. * AI security testing, monitoring, and enforcement are integrated into development and production workflows. * Security automation reduces manual effort while improving visibility and protection across AI systems. * AI security enables engineering teams to innovate quickly without compromising security, privacy, or reliability., I acknowledge by clicking "Submit Application", that the information provided is true and correct. I also understand that any willful dishonesty may render for refusal of this application or immediate termination of employment. By providing your information, you acknowledge that you have read our Zip Applicant and Candidate Privacy Notice and authorize Zip to process your data subject to those terms. Zip participates in the federal government's E-Verify program. #LI-Remote