Remote AI Image & Video Evaluator — Spot Flaws

Outlier; Charleston (WV); À distance; USD 20 000 - 29 000; Temps partiel

Avantages offerts par ce poste

Earn up to $18 USD/hr

Flexible, remote work

Free access to leading LLMs

Résumé du poste

Outlier is seeking detail-oriented English speakers to review AI-generated media remotely on a flexible schedule. You will engage in live conversations with AI voice models, compare image and video outputs, and explain what’s off in writing.

Your evaluations help AI systems improve in real-world use cases. Ideal candidates are fluent in English, highly detail-conscious, and comfortable making clear, evidence-based judgments when multiple options have flaws.

Qualifications

• Fluent English with clear written communication.
• Strong attention to detail and ability to replay or zoom in when something seems off.
• Experience writing concise, evidence-based explanations or reviews.
• Capable of thinking on your feet and explaining complex observations clearly.

Responsabilités

• Hold live conversations with AI voice models and rank naturalness, usefulness, and clarity.
• Compare AI-generated images and videos and judge which follows the prompt better.
• Inspect outputs closely to spot distortions, flicker, mismatches, or audio-visual misalignments.
• Write a specific English explanation of what you saw or heard, pointing to concrete issues.

Connaissances

Fluent English

Attention to detail

Clear written communication

Conversational English

Evidence-based explanations

Decision making under ambiguity

Formation

Bachelor's degree or above

Description du poste

Outlier is seeking detail-oriented English speakers to review AI-generated media remotely on a flexible schedule. You will engage in live conversations with AI voice models, compare image and video outputs, and explain what’s off in writing.

Your evaluations help AI systems improve in real-world use cases. Ideal candidates are fluent in English, highly detail-conscious, and comfortable making clear, evidence-based judgments when multiple options have flaws.