Remote AI Evaluation Specialist (Flexible Hours)

Outlier (by Scale AI); City Of London; Hybrid; GBP 13.000 - 18.000; Vollzeit

Benefits dieser Stelle

Flexible schedule

Free access to leading LLMs

Global expert network

Zusammenfassung

Outlier (by Scale AI) is seeking detail-oriented contributors to review AI-generated media. You will engage in live conversations with AI voice models, evaluate and rank outputs, and write concise, evidence-based explanations in English.

Tasks include comparing images and videos to the prompt, and spotting issues like distortion or misalignment between audio and video. You will work remotely on your own schedule as a freelancer, with payments processed weekly.

Qualifikationen

• Fluent English with strong written communication.
• Excellent attention to detail and ability to replay or zoom in to inspect outputs.
• Experience writing short, evidence-based explanations or reviews.
• Ability to make clear decisions when both options have flaws.

Aufgaben

• Hold live, scenario-based conversations in English with AI voice models and rate naturalness, usefulness and clarity.
• Compare AI-generated images and videos and judge which follows the prompt better.
• Inspect outputs closely to spot issues like distortion, flicker, or misalignment between audio and video.
• Provide written explanations of what looked or sounded wrong and why.

Kenntnisse

Fluent English

Attention to detail

Clear written communication

Experience writing concise reviews

Decision making under uncertainty

Ausbildung

Bachelor's degree or above

Jobbeschreibung

Outlier (by Scale AI) is seeking detail-oriented contributors to review AI-generated media. You will engage in live conversations with AI voice models, evaluate and rank outputs, and write concise, evidence-based explanations in English.

Tasks include comparing images and videos to the prompt, and spotting issues like distortion or misalignment between audio and video. You will work remotely on your own schedule as a freelancer, with payments processed weekly.