[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"job-ai-generalist-trainer-jobs-wiesbaden-germany":3},{"Ques":4,"Slug":22,"Header":23,"job_category":45},{"title":5,"content":6},"Frequently Asked Questions",[7,10,13,16,19],{"A":8,"Q":9},"Yes. This is a remote, full-time role, and you must be based in Germany.","Is this role remote?",{"A":11,"Q":12},"You will perform large language model evaluation tasks including evaluation, ranking and comparisons (RLHF-style), QA evaluation, validation of labels, prompt evaluation, content safety labeling, and writing rationales to support training data quality and model performance improvement.","What tasks will I do?",{"A":14,"Q":15},"AI experience is helpful but not required. We value strong analytical skills, attention to detail, and the ability to follow annotation guidelines compliance requirements; training is provided for project-specific rubrics and workflows.","Do I need AI experience?",{"A":17,"Q":18},"Fluency in both English and German is required, including strong reading and writing skills in both languages.","What languages are required?",{"A":20,"Q":21},"Domains vary by project and may include general knowledge, writing quality, reasoning, customer-style queries, and safety-related scenarios. The focus remains on bilingual evaluation, training data quality, and reliable judgments for RLHF and large language model evaluation.","What domains are covered?","ai-generalist-trainer-jobs-wiesbaden-germany",{"desc":24,"title":25,"content":26},"Rexzone is hiring Germany-based, bilingual (English\u002FGerman) AI Generalist Trainers to support large language model evaluation through RLHF, ranking, QA evaluation, and training data quality improvements across real-world domains.","Germany-Based English & German AI Generalist Trainer 2026 May",[27,30,33,36,39,42],{"h2":28,"desc":29},"About the Role","As a Germany-based English & German AI Generalist Trainer at Rexzone, you will evaluate and improve AI\u002FLLM workflows by assessing, ranking, and validating model-generated outputs. Your work supports RLHF, large language model evaluation, and model performance improvement by ensuring training data quality through consistent, detail-oriented judgments and clear rationales. You will follow annotation guidelines compliance standards, perform prompt evaluation and QA evaluation, and contribute to safer, more accurate model behavior across bilingual (German\u002FEnglish) content.",{"h2":31,"desc":32},"Key Responsibilities","Evaluate model outputs for correctness, relevance, helpfulness, and safety in English and German; rank and compare multiple responses to the same prompt to support RLHF; write concise, evidence-based rationales that capture reasoning and decision criteria; perform QA evaluation on labeled tasks to ensure training data quality and annotation guidelines compliance; validate edge cases, ambiguous prompts, and bilingual nuances using consistent judgment; identify labeling errors, inconsistencies, and potential policy violations, escalating as needed; contribute to prompt evaluation, content safety labeling, and training data quality checks to drive model performance improvement; document decisions and follow project-specific annotation guidelines while meeting productivity and quality targets.",{"h2":34,"desc":35},"Basic Qualifications","Based in Germany and authorized to work as a remote contractor\u002Femployee as applicable; fluent in English and German (reading and writing) with strong bilingual comprehension; strong analytical skills with the ability to compare alternatives, detect subtle issues, and justify decisions; excellent attention to detail and consistency when applying annotation guidelines; comfortable working independently in a remote environment and meeting deadlines; able to handle sensitive or safety-related content as part of content safety labeling and QA evaluation.",{"h2":37,"desc":38},"Preferred Qualifications","Prior experience in data labeling, prompt evaluation, QA evaluation, or large language model evaluation; familiarity with RLHF concepts, ranking tasks, and rubric-based assessment; experience applying annotation guidelines compliance processes and improving training data quality; understanding of common LLM failure modes (hallucinations, bias, instruction-following issues); self-driven, reliable, and proactive in raising issues and suggesting process improvements.",{"h2":40,"desc":41},"Compensation","USD $35–$40 per hour (hourly), based on experience and project scope. Remote, full-time workload.",{"h2":43,"desc":44},"How to Apply","Apply to Rexzone with a brief summary of your bilingual (English\u002FGerman) background, any AI\u002Fannotation experience, and your availability. Qualified candidates may be asked to complete a short evaluation task focused on ranking, reasoning, and annotation guidelines compliance.","AI Data Operations"]