Top AI Jobs in Singapore

Evaluate AI-generated responses for cultural appropriateness and safety using a structured rubric. Perform two independent evaluations per prompt-image pair, provide concise English rationales, join calibration sessions, support arbitration for disagreements, and meet quality and throughput targets during a 2–3 week freelance engagement (20–40 hours/week).

Evaluate AI-generated responses for cultural safety and appropriateness using a structured rubric across ~1,000 prompt-image pairs. Complete two independent evaluations per item, provide concise English rationales, attend calibration sessions, support arbitration, and meet quality and throughput targets over a 2–3 week freelance engagement.