Remote AI Safety Red Teamer — Frontier Model Testing

Mercor is seeking an AI Safety Specialist to remotely stress-test frontier AI models by designing adversarial prompts and evaluating unsafe behaviors. The role emphasizes documentation, collaboration with researchers, and contributing to safety benchmarking. Ideal candidates bring 5 years in AI safety or red teaming, plus strong analytical and written communication skills. This is a remote contract position offering competitive hourly compensation. J-18808-Ljbffr