About this role
Role Overview
Help improve the safety of advanced AI models by applying Belgian Dutch language expertise and cultural judgment to sensitive topics. You will evaluate and strengthen how models respond in Dutch as spoken in Belgium. No prior AI or machine learning experience is required, and training on the workflow will be provided.
Key Responsibilities
- Create expert-level prompts in Belgian Dutch covering a range of sensitive subject areas.
- Use structured guidelines to classify prompts and conversations.
- Identify adversarial wording and escalation patterns, documenting the reasoning behind each judgment.
- Use language fluency and cultural context to assess and improve model handling of sensitive Belgian Dutch content.
Qualifications
- Native or near-native fluency in Dutch as spoken in Belgium, with business-level written English.
- Currently based in Belgium. This position focuses specifically on Belgian Dutch; Netherlands Dutch is covered separately.
- Bachelor''s degree completed or in progress.
- Strong written reasoning, close attention to detail, and sound judgment regarding sensitive and dual-use information.
Preferred Experience
- Experience reviewing, grading, or red-teaming written or technical content.
- Background in trust and safety, content moderation, policy evaluation, or adversarial testing.
Work Terms
- Hourly engagement based in Belgium.
- Immediate start.
Compensation
48 to 52 per hour.