Welo Global
Project Perseus | Data Labeling Associate - Marathi Speakers (Human-in-the-Loop AI)
California +9 · On-site · Full-time · Posted Oct 1, 2026
- Location
- California
Sunnyvale, CA
San Francisco, CA
Seattle, WA
Redmond, Washington
Burlingame, CA
New York City Metropolitan Area
Los Angeles, CA
Washington, D.C.
Boston, MA - Experience
- 1–2 years of professional writing experience; 2+ years of relevant experience with safety considerations in GenAI data delivery.
- Education
- University degree (Bachelor’s or higher).
- Visa sponsorship
- Must be authorized to work in the U.S.; no visa sponsorship.
About the job
Evaluates AI model outputs and data, identifies quality issues and performance trends, and provides feedback to improve evaluation frameworks and model behavior.
What you'll do
- Review AI outputs and data for patterns, edge cases, and failure modes, and provide structured feedback.
- Identify performance trends and document clear, actionable findings.
- Contribute to evaluation frameworks, quality standards, team calibrations, and stakeholder discussions.
What they ask for
- Native-level Marathi proficiency and a university degree (Bachelor’s or higher).
- B2 or superior English; 1–2 years of professional writing experience.
- 2+ years of relevant experience with safety considerations in GenAI data delivery.
- Strong critical thinking, attention to detail, and ability to apply complex writing rules consistently.
Nice to have
- Experience in data quality, QA, annotation, or analysis.
Pay
USD 34–34 per-hour-wage; base salary status not stated.
AI focus
Evaluates AI model outputs and behavior, identifies gaps and failure modes, and helps improve evaluation frameworks and model performance.
Model-written summary · the original posting is authoritative