Skip to content
All open roles
AssuranceUSA / Remote4 to 10 years

Security Engineer, Model Threats

Red-team our own products before anyone else does. Prompt injection, tool abuse, data exfiltration through retrieval: find it here, not in a client environment.

What you will do

Build and run adversarial test suites against our agents and retrieval pipelines.

Work with engineering on mitigations that survive the next model swap.

Own threat modelling for new product surfaces.

Support client security reviews with honest findings and clear remediation.

What we are looking for

Offensive backgroundApplication or cloud security testing experience, ideally with LLM systems.
EngineeringYou write the tooling you need rather than filing a ticket for it.
AI systemsUnderstanding of how retrieval, tool calling and agent loops actually fail.
ReportingFindings written so an engineer can fix them without a meeting.

Technology stack

TestingAdversarial prompt suites · Burp · Custom harnesses
CloudAWS · Azure · Kubernetes security
LanguagesPython · Go
Send a CV and a short note on something you built and had to keep running to jobs@zitrino.com. Put the role title in the subject line. We read every application ourselves and reply either way, usually within a week.