AI Safety Red Team Tester — Remote, Adversarial Prompting

Alignerr Corp.; España; Remote; EUR 34,000 - 83,000; Part time

Benefits offered by this job

Flexible schedule

Remote work

Contract-based projects

Job summary

Alignerr is seeking AI Red Team Testers to probe, prompt, and challenge AI systems to uncover blind spots before they cause harm. This remote, hourly contract role lets you design tests and push models to their limits without a security background.

You will document failures with reproducible steps, rate severity, and contribute detailed reports to safety researchers. Work asynchronously with global engineers, set your own schedule, and choose engaging, task-based projects.

Qualifications

• Naturally curious and creative thinker who enjoys puzzles and edge cases.
• Comfortable thinking adversarially and looking for what could go wrong.
• Systematic and detail-oriented in documenting findings.

Responsibilities

• Design imaginative prompts and scenarios to expose AI weaknesses.
• Attempt to elicit incorrect, unsafe, biased, or inappropriate outputs from AI models.
• Document every failure mode with clear, reproducible steps.
• Rate the severity and impact of identified issues.
• Contribute structured, detailed reports to inform AI safety research teams.
• Collaborate asynchronously with safety engineers and researchers worldwide.

Skills

Curiosity or Creativity

Adversarial thinking

Problem solving

Strong written communication

Job description

Alignerr is seeking AI Red Team Testers to probe, prompt, and challenge AI systems to uncover blind spots before they cause harm. This remote, hourly contract role lets you design tests and push models to their limits without a security background.

You will document failures with reproducible steps, rate severity, and contribute detailed reports to safety researchers. Work asynchronously with global engineers, set your own schedule, and choose engaging, task-based projects.