Remote AI Security Tester - Red Team & Prompts

Alignerr Corp.; New York (NY); A distancia; USD 90.000 - 165.000; A tiempo parcial

Ventajas ofrecidas por este puesto de trabajo

Remote work

Flexible schedule

Long-term contract potential

Descripción de la vacante

Alignerr is seeking security-minded professionals to stress-test and harden AI systems by finding vulnerabilities before bad actors do. This is an hourly contract, remote, with a commitment of 10–40 hours per week.

You will conduct red-team exercises, craft adversarial prompts, and deliver clear, actionable reports that inform engineering teams and help strengthen guardrails across large language models. The role emphasizes independent, task-based work and staying current on evolving AI security

Formación

• Solid understanding of cybersecurity concepts, threat modeling, or penetration testing.
• Hands-on experience with AI/ML systems or prompt engineering.
• You are a creative, analytical thinker who enjoys finding cracks in complex systems.
• You write clearly and document findings in an organized, structured way.
• You’re comfortable working independently on task-based, asynchronous assignments.
• Familiarity with open-source AI platforms is a plus.
• Background in infosec or AI safety research is a bonus — but not required.

Responsabilidades

• Conduct red-teaming exercises to identify security weaknesses and failure modes in AI systems.
• Craft adversarial prompts and edge-case scenarios designed to probe and stress-test model guardrails.
• Evaluate AI outputs for safety risks, bias, and policy compliance.
• Document vulnerabilities, exploits, and unexpected behaviors in clear, structured reports.
• Collaborate with engineering teams to recommend practical mitigations and improvements.
• Stay current on emerging AI security threats, jailbreak techniques, and evolving best practices.
• Help define and refine security evaluation rubrics and testing protocols.

Conocimientos

Cybersecurity concepts

Threat modeling

Penetration testing

AI/ML systems

Prompt engineering

Adversarial thinking

Clear reporting

Herramientas

Open‑source AI platforms

Descripción del empleo

Alignerr is seeking security-minded professionals to stress-test and harden AI systems by finding vulnerabilities before bad actors do. This is an hourly contract, remote, with a commitment of 10–40 hours per week.

You will conduct red-team exercises, craft adversarial prompts, and deliver clear, actionable reports that inform engineering teams and help strengthen guardrails across large language models. The role emphasizes independent, task-based work and staying current on evolving AI security