Full job description

Our vision is to transform how the world uses information to enrich life for all.

Micron Technology is a world leader in innovating memory and storage solutions that accelerate the transformation of information into intelligence, inspiring the world to learn, communicate and advance faster than ever.

AI agents are becoming part of how work gets done, and adversaries are learning to attack them and to attack with them. This role exists to protect those systems and to make sure we can detect, contain, and recover when they are targeted. As a Staff AI Security Engineer, you build the tools, solutions, architectures, and frameworks that secure our sanctioned AI estate and defend against agentic threats across their full lifecycle, from prevention through response and recovery. This is a hands-on, individual-contributor role on the Cyber Defense and Resiliency team, working with the Senior Manager, Technical Network Security. You spend most of your time building, and your work sets the technical direction that others follow.

You operate where emerging AI capability meets practical security engineering! On one side you build and operate the protective controls that keep our approved AI solutions safe to use. On the other you translate novel agentic attack patterns into working defenses, automated response actions, and reusable frameworks that the broader organization can adopt. You are comfortable moving fast in an area where the playbooks are still being written, and you bring rigor to a space that rewards it.

Responsibilities:

Build and harden security tools that protect AI agents, model-backed services, and supporting infrastructure by securing system configurations, access, and integrations.
Operate and mature AI guardrails, including in-line model protections, prompt-injection defenses, and data-loss prevention, in partnership with AI enablement and governance teams.
Design detection capabilities for agentic threats, including prompt injection, tool/function abuse, identity and credential misuse, data exfiltration, and agent memory or context manipulation.
Develop automated, machine-speed response and containment tools with tiered authority for reversible and irreversible actions.
Enable rapid, scalable revocation of compromised agent access and credentials, while restoring agent configurations, tool registries, and supporting services to a known-good state.
Create and document reusable architectures, reference designs, and frameworks that enable consistent security and defense of agentic systems across the organization.