About this role
AI Pentester & Security Risk Researcher
United States, Washington, Redmond
+1 more
Apply now Add to cart
Use AI to find out how well you match with this job
Upload resume
Job description
Company and benefits
Job number
200065916
Date posted
Nov 19, 2026
Work site 3 days / week in-office
Travel Less than 25%
Profession Software Engineering
Discipline Software Engineering
Role type Individual Contributor
Employment type Full-Time
Overview
The Trust and Integrity Protection (TrIP) team supports the company's overall security mission by providing key security services that help protect systems, services and data. We are looking for a motivated cybersecurity engineer with experience in AI penetration testing and proactive risk hunting.
Are you passionate about finding security vulnerabilities and hidden risk in enterprise-scale systems, with a specific focus on Artificial Intelligence? Do you want to find the risks that traditional reviews, attestations and scans miss? TrIP's AI Security and Application Security Team is responsible for AI penetration testing, application security and risk hunting across the entire Microsoft Customer and Partner Solutions (MCAPS) technology estate.
As a member of TrIP's AI Security and Application Security Team, you will hunt for risk proactively by connecting signals across data sources, telemetry, inventories and business processes. You will perform AI penetration testing against AI and enterprise systems, develop novel attack techniques, and turn systemic, repeating risk patterns into automated detections and agent test cases. You will also act as a trusted risk advisor, helping business and engineering teams launch solutions safely and giving leadership clear visibility into risk.
Responsibilities
-
Proactive risk indentification and discovery
-
Proactively find risks that traditional methods such as reviews, attestations and scans do not catch, across the internal, external and support portfolios.
-
Use data sources, telemetry, asset inventories and business processes to connect signals across systems, spot gaps and anomalies, and uncover hidden or emerging risk.
-
Form and test risk hypotheses, validate findings, and quantify exploitability, blast radius and business impact.
-
Drive remediation with asset owners and track risk to closure.
-
AI penetration testing and novel scenarios
-
Discover and exploit vulnerabilities end to end to assess the security of AI systems, agents, copilots, connectors and generative AI solutions.
-
Execute penetration testing operations on production AI systems using real-world adversarial tactics and techniques.
-
Research and test new and emerging threats, including prompt injection, agent-to-agent trust abuse, tool and memory poisoning and model supply chain risk, before they areexploited.
-
Stay current with penetration testing tools, methodologies, tactics and mitigations.
-
Accelerate
-
Turn systemic, repeating risk patterns into automated detections, agent test cases or policy updates, so the same class of risk is caught at scale without another manual hunt.
-
Develop tools and techniques that scale AI penetration testing, vulnerability discovery and risk hunting, including contributions to TrIP testing and assessment agents.
-
Develop, operationalize and maintain penetration testing and risk hunting procedures and methodologies. Business enablement
-
Act as a trusted risk advisor, translating technical risk into business impact and clear go, go-with-conditions or no-go guidance.
-
Partner with teams early in design to help