About The Role
Rex.zone is hiring Germany-based, bilingual (English/German) AI Generalist Trainers to support AI/LLM workflows through RLHF, large language model evaluation, and training data quality work that drives model performance improvement. You will evaluate and improve model outputs in English and German, follow detailed annotation guidelines, and write clear rationales that improve dataset quality.
About The Role
Rex.zone is hiring Germany-based, bilingual (English/German) AI Generalist Trainers to support AI/LLM workflows through RLHF, large language model evaluation, and training data quality work that drives model performance improvement. You will evaluate and improve model outputs in English and German, follow detailed annotation guidelines, and write clear rationales that improve dataset quality.
Key Responsibilities
- Evaluate and compare model-generated responses in English and German using defined rubrics
- Rank outputs for RLHF / preference data and document reasoning and trade-offs
- Perform QA evaluation on labeled datasets for consistency, correctness, and policy alignment
- Write concise rationales to support training data quality improvements
- Validate edge cases, ambiguous prompts, and multi-turn conversations
- Apply content safety labeling and policy checks; elevate uncertain cases with evidence
- Track errors, perform spot checks, and propose rubric clarifications
- Collaborate asynchronously with reviewers to maintain quality and throughput targets
Basic Qualifications
- Must be based in Germany and able to work remotely from Germany
- Fluency in English and German (C1/C2 or equivalent), including professional reading and writing
- Strong analytical skills and attention to detail
- Ability to follow detailed instructions and annotation guidelines compliance requirements
- Ability to explain decisions clearly with structured rationales and validations
- Reliable internet connection and ability to meet productivity expectations
Preferred Qualifications
- Experience with data labeling, prompt evaluation, QA evaluation, or content moderation
- Familiarity with LLM evaluation and RLHF concepts (helpful but not required)
- Experience with annotation tools and training data quality processes
Compensation
$30-50 USD per hour (full-time, remote; Germany-based requirement).
#J-18808-Ljbffr
Germany-Based English & German AI Generalist Trainer Arbeitgeber: Rex.zone
Rex.zone ist ein hervorragender Arbeitgeber, der eine dynamische und unterstützende Arbeitsumgebung in Berlin bietet. Mit einem starken Fokus auf digitale Marketingstrategien und innovativen Technologien fördert das Unternehmen die berufliche Weiterentwicklung seiner Mitarbeiter durch kontinuierliche Schulungen und spannende Projekte. Die remote-freundliche Kultur ermöglicht es den Mitarbeitern, flexibel zu arbeiten und gleichzeitig Teil eines kreativen Teams zu sein, das sich leidenschaftlich für den Erfolg seiner Kampagnen einsetzt.
