Secure AI Engineer - Red Team & Safe AI Architect

Latitude; Washington (District of Columbia); Hybrid; USD 78,000 - 176,000; Full time

Job summary

Booz Allen hires a Secure AI Engineer to combine AI, cybersecurity, and offensive security to assess, harden, and improve AI-enabled systems. You will work with clients to design secure AI architectures, evaluate resilience, and implement mitigations for safe deployment of generative and autonomous AI.

The role requires 5+ years in AI/ML, 3+ years with LLMs, and knowledge of AI security frameworks; hybrid remote work is expected.

Qualifications

• Bachelor's degree required in CS, Cybersecurity, or Engineering.
• 5+ years building, deploying, or evaluating AI/ML systems.
• 3+ years developing applications with LLMs and modern AI frameworks.
• Experience with AI red teaming, adversarial AI, or AI governance.
• Knowledge of AI security principles and risk management.

Responsibilities

• Advise clients on secure design, deployment, and governance of AI systems.
• Conduct AI red team assessments against LLMs, RAG, AI agents, and workflows.
• Identify AI attack vectors and implement mitigations to improve security.
• Build, test, and evaluate agentic AI systems and assess security posture.

Skills

AI security principles

Secure AI development

Adversarial AI

AI governance

Red teaming AI

Education

Bachelor's degree in CS, Cybersecurity, or Engineering

CISSP/OSCP/Security+ or GIAC

Tools

LangGraph

CrewAI

AutoGen

Semantic Kernel

OpenAI Agents SDK

PyRIT/Garak/Inspect AI

Job description

Booz Allen hires a Secure AI Engineer to combine AI, cybersecurity, and offensive security to assess, harden, and improve AI-enabled systems. You will work with clients to design secure AI architectures, evaluate resilience, and implement mitigations for safe deployment of generative and autonomous AI.

The role requires 5+ years in AI/ML, 3+ years with LLMs, and knowledge of AI security frameworks; hybrid remote work is expected.