Linguistics AI Evaluation Expert

Use your linguistics expertise to create challenging problems and evaluate AI-generated responses. Work asynchronously, using independent judgment to provide feedback that can improve model performance.

• Create expert-level problems based on linguistic analysis, corpus methods, and computational workflows.
• Assess responses for technical accuracy, analytical rigor, and fit with real linguistic practice.
• Write clear, structured feedback that explains weaknesses.

What it pays and takes

• Pay: Up to $80 per hour.
• Hours: 20+ hours per week. The work is part-time, flexible, and has no set schedule.
• Location: Open to candidates in the United States; US work authorization is required.
• Language: Fluency in English.
• Education: A PhD or MS in Linguistics or Computational Linguistics is requested. Candidates who meet at least two core qualifications are also encouraged to contribute.
• Research: Publications and expertise in a language or linguistic phenomenon that AI models handle poorly.
• Knowledge: Advanced knowledge in areas such as morphology, syntax, phonology, formal grammars, corpus methods, low-resource languages, phonetics, or computational and NLP tools.
• Skills: Strong writing, attention to detail, and the ability to explain complex linguistic ideas clearly.
• Helpful experience: Praat, IPA, corpus analysis software, computational or NLP tools, low-resource languages, or typologically diverse languages.

How it works

Apply on OpenTrain with your resume, then complete the application on the hiring site.

About AI training work

AI training is the human work behind AI: people create examples and assess model responses so systems can improve. This role uses specialist linguistics knowledge to judge responses and give feedback on weak analysis.