Agentic AI Security Engineer

LTS

6d•Remote

About The Position

LTS is seeking a highly skilled Agentic AI Security Engineer to ensure our AI systems are secure, trustworthy, resilient, and governed responsibly. You'll design safeguards that protect AI agents, retrieval pipelines, prompts, tools, memory, and model interactions from misuse while ensuring every AI response remains explainable, auditable, and aligned with enterprise security and compliance requirements. This role focuses on securing AI systems, not simply securing infrastructure. The Agentic AI platform is designed to help engineers understand, analyze, and modernize one of the most consequential legacy software systems still operating today. Our platform enables engineers to ask questions in plain English and receive explainable, verifiable answers traced directly back to decades of production source code. Rather than replacing engineers, we're building AI that accelerates engineering through transparency, traceability, and intelligent reasoning. We're building an AI-native engineering platform supporting the modernization of mission-critical healthcare systems serving millions of Veterans nationwide. The platform is designed for deployment across federal enterprise environments and is being engineered to align with FedRAMP security controls, Zero Trust principles, and federal compliance requirements. You'll work alongside AI architects, software engineers, platform engineers, cybersecurity engineers, and product leaders to embed AI security into every stage of development from experimentation through production deployment. We’re building an AI platform where security, transparency, and trust are foundational, not afterthoughts.

Requirements

Bachelor's degree in Computer Science, Cybersecurity, Artificial Intelligence, Software Engineering, Information Security, or a related technical discipline (or equivalent professional experience).
7+ years of software engineering, cybersecurity engineering, AI engineering, or application security experience.
Experience designing secure cloud-native or distributed software systems.
Experience with Large Language Models (LLMs), AI applications, or Agentic AI platforms.
Experience securing APIs, microservices, and enterprise applications.
Knowledge of OWASP Top 10 and secure software development practices.
Understanding of AI-specific security risks including: Prompt injection, Jailbreaking, Data poisoning, Model abuse, Adversarial inputs, Hallucination mitigation, Sensitive data leakage
Experience with cloud security across AWS, Azure, or Google Cloud.
Strong programming experience in Python or TypeScript.
Excellent communication and collaboration skills.
A mindset of both a security engineer and a software engineer.
Experience with solving problems that don’t yet have established playbooks.
Ability to stay current with emerging AI threats and defensive techniques.
Experience with ownership of complex technical challenges.
Ability to collaborate effectively across engineering disciplines.
Capability to influence how secure AI systems are built in highly regulated environments.
Strong ability to balance innovation with responsible engineering.

Nice To Haves

Experience securing Retrieval-Augmented Generation (RAG) systems.
Experience with AI guardrails and policy engines.
Familiarity with frameworks such as LangGraph, LangChain, CrewAI, Semantic Kernel, AutoGen, or LlamaIndex.
Experience with vector databases including Pinecone, Weaviate, Qdrant, Milvus, or Azure AI Search.
Experience implementing AI observability, evaluation, or monitoring solutions.
Experience with NIST AI Risk Management Framework (AI RMF), Responsible AI practices, or AI governance frameworks.
Experience supporting Federal Government or healthcare environments.
Familiarity with Zero Trust Architecture and secure DevSecOps practices.
Professional certifications such as CISSP, CCSP, Security+, GIAC, or cloud security certifications are a plus.

Responsibilities

Design and implement security controls for autonomous and multi-agent AI systems.
Secure agent orchestration, tool execution, memory, and external integrations.
Identify and mitigate emerging AI-specific attack vectors and vulnerabilities.
Secure Retrieval-Augmented Generation (RAG) pipelines, embeddings, vector databases, and enterprise knowledge repositories.
Design controls that prevent unauthorized knowledge access, data leakage, and information exposure.
Implement secure handling of sensitive enterprise and healthcare data throughout AI workflows.
Perform threat modeling for AI applications, agent architectures, prompts, APIs, and retrieval systems.
Assess risks associated with prompt injection, jailbreak attempts, indirect prompt attacks, tool misuse, hallucinations, data poisoning, model abuse, and adversarial inputs.
Develop mitigation strategies that reduce AI-specific security risks while maintaining usability.
Build guardrails that improve trustworthy AI behavior.
Design policy enforcement, human-in-the-loop approval workflows, content filtering, and AI governance mechanisms.
Help define organizational standards for responsible AI development and deployment.
Design monitoring capabilities that detect abnormal agent behavior, misuse, prompt manipulation, and anomalous model interactions.
Implement logging, traceability, and audit capabilities supporting explainability and regulatory compliance.
Build mechanisms for continuous AI risk assessment and operational visibility.
Partner with software engineers to integrate AI security into development workflows.
Conduct security reviews of AI features before production deployment.
Promote secure AI engineering practices across the product organization.