Outlier, owned and operated by Scale AI, is committed to improving the intelligence and safety of AI models by providing high-quality data for large language models (LLMs). The job aims to address the challenge of training AI systems with expert human feedback to enhance their performance and reliability.
Requirements
- Ability to develop complex, professional-level prompts and evaluate nuanced AI reasoning.
- Clear, concise, and engaging writing to explain decisions or critique responses.
- Commitment to accuracy and ability to assess technical aspects of model outputs.
- Experience in fields like literature, creative writing, history, philosophy, theology, etc. (Nice to have).
- Prior writing or editorial experience (content strategist, technical writer, editor, etc.). (Nice to have)
- Interest or background in AI, machine learning, or creative tech tools. (Nice to have)
- Fluent English language skills.
Responsibilities
- Adopt a user mindset to produce natural data to meet the realistic needs you have or would use AI for.
- Use the tool of rubrics to address user needs in a structured way.
- Evaluate AI outputs by reviewing and ranking reasoning and problem-solving responses from large language models.
- Contribute across projects depending on your specific skillset and experience.
- Develop complex, professional-level prompts and evaluate nuanced AI reasoning.
- Provide clear, concise, and engaging writing to explain decisions or critique responses.
- Assess technical aspects of model outputs with attention to detail.
Other
- Bachelors degree or higher (or currently enrolled).
- Flexible schedule with no contracts or 9-to-5 requirements.
- Most experts spend 5-10 hours/week, up to 40 hours working from home.
- Fluent English speakers only.
- Ability to work remotely and collaborate with a global community of contributors.
