About this role
Help improve advanced language models by transcribing, annotating, and evaluating Russian audio and video content. This remote role supports audio AI research and may include occasional in-person collaboration in the San Francisco Bay Area.
Key Responsibilities- Listen to, analyze, and accurately transcribe Russian audio and video while following detailed instructions.
- Create high-quality written Russian outputs, with English support work as needed.
- Capture tone, intent, formality, regional expressions, dialectal variation, and contemporary Russian usage where relevant.
- Develop Russian and English evaluation rubrics, grading guidance, and documentation for consumer audio scenarios.
- Identify Russian-specific linguistic nuances, grammatical complexities, colloquialisms, and edge cases.
- Test language-model prompts and assess outputs for accuracy, completeness, fluency, and instructional clarity.
- Provide structured feedback to improve model performance on Russian audio tasks.
- Take part in quality-review cycles and collaborate with project leads to clarify requirements and improve task design before datasets are incorporated into benchmarks.
- Native or near-native spoken and written Russian, plus professional English fluency.
- Strong writing, editing, and critical-thinking skills.
- Familiarity with spoken Russian, regional vocabulary, dialects, and current language usage.
- Ability to independently manage time, meet deadlines, and accurately analyze Russian audio in general consumer contexts.
- College student or recent graduate.
- Background in linguistics, humanities, social sciences, journalism, translation or localization, or a technical discipline.
- Experience with Russian transcription, annotation, localization, evaluation, or research workflows.
- Interest in AI, language models, or applied research.
- Remote, hourly engagement.
- Must be based in the San Francisco Bay Area.
- Commitment of 10 to 20 hours per week.
- $50 per hour.
- Complete an AI-led interview of approximately 15 minutes. Selected candidates will complete onboarding before beginning project work.