Role Overview
The AI Agent Security Architect position at Replit serves as the primary technical authority for autonomous and AI agent security frameworks. This full-time remote-friendly role is based out of Foster City, California, and suits seasoned security professionals with a strong background in artificial intelligence and machine learning protection.
Key Responsibilities
Define long-term vision and architectural patterns for securing autonomous agent workflows, Model Context Protocol integrations, and multi-agent coordination.
Architect and deploy dynamic input/output guardrail systems, semantic firewalls, and real-time intent verification filters.
Design secure, short-lived micro-isolated environments in partnership with infrastructure teams where agents can safely execute code and run shell commands.
Conduct specialized threat modeling against non-deterministic systems and lead automated and manual AI red-teaming initiatives.
Build fine-grained permission models and step-up Human-in-the-Loop authorization patterns for agents acting on behalf of users.
Design strict data isolation and poisoning prevention controls for vector databases and retrieval-augmented generation pipelines.
Maintain authoritative secure design patterns and SDKs for engineering teams building internal or customer-facing AI agent capabilities.
Identify, quantify, and document non-deterministic and agentic security risks within the Cybersecurity Risk Register.
Design tamper-evident logging and telemetry standards for agent reasoning chains and tool execution history.
Translate complex AI security controls into enterprise-ready audit documentation and support sales inquiries regarding AI safety.
Requirements
Six or more years of experience in security engineering or security architecture.
At least two years of dedicated focus on AI/ML security, large language model application security, or securing agentic frameworks.
Deep understanding of agentic architectures and protocols like Model Context Protocol, LangChain, AutoGen, and vector database technologies.
Hands-on expertise with threat vectors unique to agentic artificial intelligence, including indirect prompt injection, goal hijacking, and tool parameter tampering.
Strong background in applying safety standards and risk frameworks such as OWASP Top 10 for LLMs and AI Agents, NIST AI RMF, and MITRE ATLAS.
Proficiency in Python, Go, TypeScript, or JavaScript alongside code review capabilities.
Experience creating and evangelizing technical security architecture documentation.
Exceptional ability to communicate complex non-deterministic risks to both engineers and executives.
Nice to Have
Experience designing runtime sandboxing and isolation technologies such as Firecracker, gVisor, Docker, or WebAssembly for dynamic code execution environments.
Skills and Tools
Python
Go
TypeScript
JavaScript
Docker
React
LLMs
LangChain
RAG
Model Context Protocol
Salary and Benefits
Competitive salary and equity packages.
Four hundred one k program with a four percent match available for United States employees.
Comprehensive health, dental, vision, and life insurance coverage.
Short-term and long-term disability protection.
Paid parental, medical, and caregiver leave.
Flexible time off alongside standard holidays.
Commuter benefits for in-office personnel in the United States.
Monthly wellness stipend.
Autonomous work environment.
In-office setup reimbursement and office amenities.
Quarterly team gatherings.
About the Company
Replit operates an agentic software creation platform designed to let anyone build applications utilizing natural language. Serving millions of users worldwide, the organization removes traditional barriers to application development to democratize software creation.
How to Apply
Please visit the official company careers page to submit your application for the AI Agent Security Architect position.
