AI Safety Red Team Tester — Remote & Flexible

Alignerr is seeking AI Red Team Testers to systematically probe and outwit cutting-edge AI systems, uncovering blind spots before they cause real harm. No security background required—just a curious, adversarial mindset and clear written communication.

This fully remote, hourly contract role offers 10–40 hours per week with ongoing opportunities for project-based work. You will design prompts, reveal weaknesses, document failure modes, and collaborate with research teams to improve model safety.