AI Evaluation Researcher - Remote & Flexible (Freelance)

Outlier Ai; Birmingham; A distancia; GBP 13.000 - 18.000; A tiempo parcial

Ventajas ofrecidas por este puesto de trabajo

Flexible schedule

Remote work from UK

Weekly payments

Descripción de la vacante

Outlier, a Scale AI–backed AI training platform, is seeking detail‑oriented English speakers to review AI‑generated media and rate how well outputs follow prompts. You’ll roleplay live conversations with an AI voice model and explain in writing what looked or sounded off, all on your own schedule from anywhere in the UK.

Typical tasks include comparing AI-generated images and videos, rating adherence and realism, and providing concise, evidence‑based explanations in English.

Formación

• Fluent English speaker with strong, clear written communication.
• Excellent attention to detail and ability to replay or zoom in to spot issues.
• Ability to think on your feet and discuss realistic scenarios.
• Experience writing short, evidence-based explanations or reviews.
• Pursuing or holding a Bachelor's degree or above is preferred.

Responsabilidades

• Holding a live, scenario-based conversation in English with AI voice models and ranking naturalness.
• Comparing pairs of AI-generated images/videos and judging adherence to prompts.
• Inspecting outputs closely to spot problems like distorted anatomy or misaligned audio.
• Writing concise explanations in English that point to observed issues.

Conocimientos

Fluent English

Detail oriented

Clear written communication

Evidence-based explanation writing

Decision making in ambiguous scenarios

Educación

Bachelor's degree or above

Descripción del empleo

Outlier, a Scale AI–backed AI training platform, is seeking detail‑oriented English speakers to review AI‑generated media and rate how well outputs follow prompts. You’ll roleplay live conversations with an AI voice model and explain in writing what looked or sounded off, all on your own schedule from anywhere in the UK.

Typical tasks include comparing AI-generated images and videos, rating adherence and realism, and providing concise, evidence‑based explanations in English.