About Rex.zone Rex.zone is hiring Germany-based AI Generalist Trainers to support large language model evaluation through RLHF, LLM evaluation, and training data quality workflows. You will assess, rank, and QA model outputs in English and German, write clear rationales, and follow annotation guidelines compliance to drive model performance improvement.
About The Role As a Germany-based English \& German AI Generalist Trainer at Rex.zone, you will improve AI systems by performing large language model evaluation across diverse domains including general knowledge, customer-support style tasks, summarization, rewriting, reasoning, and content safety labeling.
Responsibilities
- Evaluate model-generated responses for correctness, completeness, and helpfulness in English and German
- Rank multiple outputs using RLHF-style preference criteria and rubric-driven scoring
- Perform QA evaluation and validation checks to ensure training data quality and annotation guidelines compliance
- Write concise, well-structured rationales explaining reasoning behind rankings and edits
- Identify ambiguity/edge cases, escalate guideline gaps, and propose clarifications
- Review prompts and responses for policy alignment, including content safety labeling when required
- Track errors, patterns, and feedback to support continuous model performance improvement
Basic Qualifications
- Based in Germany (remote role) and eligible to work
- Fluent in German and English with strong writing skills in both languages
- Strong analytical skills, attention to detail, and consistency following guidelines
- Comfortable with web tools, spreadsheets, and structured review workflows
Preferred Qualifications
- Experience with data labeling, prompt evaluation, QA evaluation, or LLM evaluation
- Familiarity with RLHF concepts, rubric-based scoring, and training data quality best practices
- Experience with editorial review or policy-aligned content review
How To ApplyApply with a brief summary of your bilingual English/German experience, your location in Germany, and any relevant evaluation/annotation/QA background. Qualified candidates may complete an assessment focused on reasoning and annotation guidelines compliance.
