This is a paid, in-person user research study about how people use their phones, apps and everyday digital tools, with attention to how habits differ across languages and cultures. It is aimed at native French speakers living in New York who can take part in a short multi-day session with a research team.
Pay not disclosed
The platform does not publish the pay for this job.
Open to residents of: United States.
You apply on Mindrift, the AI training platform run by Toloka: submit your application for the role, pass a qualification step, then work on projects that match your expertise.
Official posting: https://apply.workable.com/j/E29FDB07D0
Last checked on October 14, 2026.
This role involves evaluating and improving AI systems designed for Chinese language learning by assessing the linguistic accuracy, naturalness, and educational quality of AI-generated speech and text, as well as learner output across proficiency levels. The position is ideal for Chinese language experts with teaching or evaluation experience who want to directly influence how AI understands and teaches Chinese.
This role involves evaluating and improving AI systems trained on Italian-language content by assessing the linguistic accuracy and naturalness of AI-generated speech and text, as well as evaluating learner language across proficiency levels. The position is ideal for Italian language experts with teaching experience who want to shape how AI understands and teaches Italian.
This role involves evaluating and improving Arabic translations and AI-generated text by assessing fluency, tone, and cultural authenticity. Native or near-native Arabic speakers will teach language models to produce more natural, human-like Arabic output by rating translations, identifying gaps that automated systems miss, and rewriting model outputs to match native speaker standards.
A Bengali specialist evaluates and corrects AI-generated Bengali text, focusing on accurate pronoun usage, regional variations, and natural phrasing. The role involves reviewing translations, scoring model outputs, and providing native-speaker feedback to improve AI language understanding.
This role involves editing and quality-checking Brazilian Portuguese voice recordings so they can be used to train text-to-speech and speech AI systems. The work centers on cleaning audio, checking technical standards, and judging pronunciation and naturalness at high volume. It suits audio engineers with native or near-native Brazilian Portuguese who have experience with speech recordings.
Posted October 13, 2026
This is an in-person research study on how native German speakers use their phones, apps and everyday digital tools. Participants spend one to three days with a research team, walking through their habits, sharing examples of their daily phone activity and filling in a short questionnaire. It is aimed at German speakers who live locally and can commit to consecutive days.
Posted October 13, 2026
