Amazon's Security team is looking for a Senior Agentic AI Engineer to design and build agent-based systems that enable the security team to operate at scale on a first-of-its-kind AI Factory. This is a greenfield opportunity: we are building an entirely new security program on a novel platform with no inherited tooling or automation. We expect our teams to be building agent-first, using teams of agents for delivery across detection, response, compliance, and operational workflows. You will architect and implement multi-agent systems that decompose complex security processes into coordinated autonomous workflows. We value strong process decomposition skills, production-quality engineering, and the ability to design systems that operate reliably in disconnected, high-security environments.
This position requires that the candidate selected be a US Citizen and must currently possess and maintain an active TS/SCI security clearance with polygraph.
Key job responsibilities
• Architect and build multi-agent systems for security operations, detection, compliance, and workflow automation.
• Decompose complex security processes into coordinated agent workflows.
• Design reliable agent orchestration patterns for high-security, air-gapped environments.
• Implement robust error handling, observability, and human-in-the-loop controls for autonomous systems.
• Define standards and patterns for agent development across the security team.
• Evaluate and integrate agentic frameworks, adapting them for classified environments.
• Partner with security engineers to identify processes suitable for agent-based automation.
A day in the life
You will own security-centric AI outcomes: setting technical direction, identifying the highest-value opportunities, defining measurable success criteria, and ensuring that agentic capabilities improve security and engineering results rather than simply increase automation. You will work across developers, security engineers, and other technical partners to understand existing development processes and determine which work can be safely and effectively automated with agentic AI. You will distinguish those opportunities from problems better solved through conventional software or deterministic automation, build alignment around priorities, and drive selected initiatives from concept through deployment and adoption.
You will prototype agentic workflows and lead other engineers in turning successful prototypes into maintainable production capabilities. This includes defining architecture and engineering standards, decomposing work across agents and traditional components, providing agents with appropriate context and tools, establishing permissions and human approval boundaries, and ensuring the resulting automation has meaningful tests, operational measures, and accountable owners.
BASIC QUALIFICATIONS
- 6+ years of non-internship professional software development experience
- 6+ years of programming with at least one software programming language experience
- 5+ years of leading design or architecture (design patterns, reliability and scaling) of new and existing systems experience
- Experience as a mentor, tech lead or leading an engineering team
- Demonstrated experience building applications or workflows using large language models, including tool use, structured outputs, retrieval, or agentic execution.
- Current, active US Government Security Clearance of TS/SCI with Polygraph
PREFERRED QUALIFICATIONS
- 5+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations experience
- Bachelor's degree in computer science or equivalent
- Experience in written and verbal communication with the ability to present complex technical information in a clear and concise manner to executives and non-technical leaders
- Experience designing or operating production agent teams or multi-agent systems, including role specialization, delegation, routing, shared context, coordination, conflict resolution, and termination conditions.
- Experience with one or more agentic frameworks or protocols, such as LangGraph, Semantic Kernel, AutoGen, CrewAI, Strands Agents, the Model Context Protocol, or equivalent internally developed systems.
- Experience building agent harnesses that combine instructions, policies, tools, retrieval, memory, permissions, execution environments, and evaluation mechanisms.
- Knowledge of security threats specific to LLM and agent systems, including prompt injection, indirect prompt injection, tool misuse, confused-deputy behavior, excessive agency, sensitive-data disclosure, memory poisoning, supply-chain compromise, and evaluation manipulation.
- Experience creating tests and evaluations for nondeterministic systems, including quality criteria, representative test cases, failure analysis, regression testing, and human review.
- Experience operating or troubleshooting software through logs, metrics, traces, alerts, and structured execution records.
- Ability to evaluate architectural trade-offs involving security, reliability, latency, cost, scalability, and maintainability.
- Ability to rapidly prototype solutions and turn successful prototypes into maintainable, documented engineering systems.
The base salary range for this position is listed below. Your Amazon package will include sign-on payments and restricted stock units (RSUs). Final compensation will be determined based on factors including experience, qualifications, and location. Amazon also offers comprehensive benefits including health insurance (medical, dental, vision, prescription, Basic Life & AD&D insurance and option for Supplemental life plans, EAP, Mental Health Support, Medical Advice Line, Flexible Spending Accounts, Adoption and Surrogacy Reimbursement coverage), 401(k) matching, paid time off, and parental leave. Learn more about our benefits at https://amazon.jobs/en/benefits.
USA, MD, Jessup - 168,100.00 - 227,400.00 USD annually
USA, VA, Arlington - 168,100.00 - 227,400.00 USD annually
USA, VA, Herndon - 168,100.00 - 227,400.00 USD annually
