Staff Security Engineer - AI

At LinkedIn, our approach to flexible work is centered on trust and optimized for culture, connection, clarity, and the evolving needs of our business. The work location of this role is hybrid, meaning it will be performed both from home and from a LinkedIn office on select days, as determined by the business needs of the team.

This location is hybrid in our Sunnyvale office location.

About the team

LinkedIn's members entrust us with their information every day and we take their security seriously. Our core value of putting our members first powers all the decisions we make, including how we manage and protect the data of our members and customers. Information Security at LinkedIn is dedicated to protecting and securing business-critical member data and company assets. Our core mission is to empower LinkedIn to create a secure and thriving platform for every member of the global workforce. The Product Security team strives to proactively safeguard our products, applications, and infrastructure by identifying, assessing, and mitigating security and privacy risks. We are dedicated to protecting our members by researching evolving threats and attack vectors, identifying vulnerabilities, and providing security consultation to minimize potential risks. We invest significantly in automation and focus on high impact engineering projects that detect security risks. 

About the role:

As a Staff Security Engineer on the Platform & Product Security team, you will strengthen the security posture of LinkedIn's AI applications, infrastructure, and development workflows — researching emerging threats, leading threat models, designing and implementing secure-by-default frameworks, and building scalable tooling to detect and mitigate vulnerabilities across the company. 

Responsibilities:

• Lead and execute adoption of security controls across AI/ML infrastructure, including model scanning, model signing and provenance, and ML pipeline security (MLOps/MLSecOps)

• Architect and build security guardrails and paved roads for agentic workflows: hardening agent SDKs, tool access and permissioning, input/output validation, execution boundaries, sandboxing, and auditability of agent actions

• Build automation and red-team/detection tooling that continuously identifies, reproduces, and mitigates AI-specific vulnerabilities at scale; partner with Detection & Response on prompt/tool-call telemetry and AI-incident playbooks.

• Lead threat modeling and adversarial red-teaming of AI systems, and maintain a LinkedIn AI threat taxonomy mapped to industry standards. Partner with security and platform teams to embed the resulting threat models and risk assessments into AI platforms, developer tooling, and production services.

• Define AI security standards, reference architectures, and policies across model development, data ingestion, training, inference, and agentic systems; drive secure adoption of AI tooling (vendors, copilots, integrations), including data-sharing and sensitive-data-in-prompt boundaries.

• Drive cross-functional architecture and engineering decisions across platform engineering, security engineering, ML/AI teams, and product/DevEx to build developer platforms, AI systems, and infrastructure that are scalable, reliable, and secure by design.