Engineering / Expert network
Security & privacy expert
Investigate how AI systems handle permissions, sensitive data, and adversarial inputs.
This is the Rubrex security & privacy expert track for future AI evaluation project matching. It is a network application, not a confirmed vacancy.
The work you could do
- Evaluate AI behavior against documented access and data boundaries.
- Develop scoped adversarial test cases in authorized environments.
- Document reproducible findings, evidence, and the limits of a test.
What you bring
- Relevant application security, privacy engineering, or security testing experience.
- A disciplined approach to authorized testing and handling confidential information.
An example of expert review
Review a fictional assistant workflow with a defined permission boundary. Identify a step that could expose restricted data, describe an authorized test, and explain what evidence would show that the boundary held or failed.
This is an illustrative task, not an open assignment or a required application exercise. Actual work depends on the project scope and your area of expertise.
What to include in your background
Choose one or two examples that show how you check work in your field. Describe the question, the evidence you used, and the reasoning behind your decision. Tell us which parts you handled yourself and where you needed input from someone else.
You can describe research, professional practice, teaching, or independent work. Use public or anonymized examples that you have permission to share. Do not upload client records, private datasets, or confidential employer material. A clear account of your method is more useful than a list of tools alone.
How we review your application
Share your resume and describe experience relevant to this track. Our team will review your background against the role requirements and let you know the outcome by email. If there is a potential project fit, we may discuss your experience or invite you to a relevant assessment.
Project scope, timing, location requirements, and compensation are confirmed before work begins. You can read about our AI evaluation services to understand how expert judgment supports a repeatable quality check. Our application data notice explains how we handle the information you submit.