Bilingual Croatian Generalist Expert — AI Safety | $38-$42/hr Remote
Overview
Your bilingual Croatian-English fluency moves into an AI safety role with immediate start and remote flexibility. You will craft prompts in Croatian across sensitive topics and apply structured rules to judge model responses. There is no need for prior machine learning experience, since full training covers the workflow. This part-time contract pays an hourly rate and suits strong writers who can document their decisions.
What You'll Do4
- 1Write expert-level prompts in Croatian tackling sensitive subject areas such as violence, self-harm, and misinformation.
- 2Apply set classification rules to appraise prompts and user dialogues, determining whether they meet safety guidelines.
- 3Detect adversarial phrasings and escalation tactics, then note the reasoning behind each flag.
- 4Produce clear written justifications that document your decisions for review by the project team.
Requirements4
- 1Native or near-native fluency in Croatian and business-level written English.
- 2A completed or in-progress bachelor's degree.
- 3Strong written reasoning skills paired with high attention to detail.
- 4Sound judgment when handling sensitive or dual-use information.
Who Should Apply
The ideal candidate is fluent in Croatian and English, holds or pursues a bachelor's, and can explain why a particular prompt violates safety rules. The role is not a fit for someone uncomfortable with sensitive subject matter or who prefers broad instructions over precise guidelines. Rejections happen mainly when applicants miss the language proficiency bar or when their written justifications are vague and unstructured. Also, applicants without any degree or currently enrolled in one are screened out early.
Salary Insight
The role pays between $38 and $42 per hour. That range is at the higher end for part-time AI feedback work, reflecting the need for native-level Croatian and careful reasoning around sensitive topics.
Location
Required Skills
Application Tip
In your application, show that you can articulate why a given prompt is problematic. Include a one-paragraph example of how you would judge a sensitive prompt and justify your decision.
See NearSkill jobs more often in your search
How your application is processed
1Application received
Your resume and details are logged the moment you apply.
2ATS + eligibility screening
We check your profile against the role’s skills, seniority, and requirements.
3Employer sees qualified profiles only
Only candidates who clear screening move forward.
Similar open positions
Explore active roles that match your skills and interests.
Mercor
VerifiedBilingual Czech Generalist Expert — AI Safety
Your bilingual Czech-English fluency drives the mission: making advanced AI models handle sensitive topics safely. You'll draft expert-level prompts in Czech and evaluate the model's responses using a structured classification system. A major part involves identifying adversarial phrasings and writing clear justifications for your choices. No previous AI/ML experience is needed; the workflow is trained from scratch. This is a part-time remote role starting immediately, with Eastern Europe time zones preferred.
SME Careers
VerifiedCroatian Language Expert for AI Training Content
As a remote Croatian Language Expert, you review content written in Croatian and craft authoritative reference material for model training. You will develop prompts and gold-standard Croatian responses across domains like general knowledge, customer support, and education, feeding AI training data. You evaluate outputs for fluency, accuracy, tone, and cultural fit, delivering precise written feedback in English and supporting translation QA and localization QA workflows. Your work strengthens Croatian language outputs and guides quality improvements across AI data pipelines.
Mercor
VerifiedCroatian PhD STEM Expert for AI Safety Evaluation
This part-time remote role uses your scientific knowledge to help make AI models safer when they respond in Croatian. You'll compose expert-level prompts across chemistry and biology, including radiological and nuclear topics, then judge the accuracy and safe handling of model-generated answers. Native or near-native Croatian and professional written English are must-haves; prior machine learning experience isn't. Your first day can be immediate, and the team trains you on the evaluation workflow.
Mercor
VerifiedBilingual Turkish Generalist Expert — AI Safety
Put your bilingual fluency to work in AI safety. This remote, part-time project involves writing Turkish prompts that explore sensitive subjects, classifying conversation types, and flagging prompts designed to trip up safety filters. The team walks you through the entire process, so prior machine learning experience isn't required. You'll earn $23–$27 per hour while contributing to safer AI responses.
Mercor
VerifiedBilingual Ukrainian Generalist Expert — AI Safety
Help make AI systems safer by bringing your Ukrainian and English fluency to content evaluation. You will write prompts and classify conversations that touch sensitive topics, applying clear guidelines to flag adversarial phrasing. The workflow is taught from scratch, so prior AI experience is not expected. Part-time remote work starts right away, suited to writers and reviewers who enjoy careful judgment under ambiguity.
Mercor
VerifiedBilingual Polish Generalist Expert — AI Safety
Your Polish and English fluency drives this part-time remote role, where you help make AI models safer. You'll write detailed prompts in Polish and apply structured rules to judge how models respond to sensitive or adversarial topics. The job doesn't need a machine learning background; the team provides complete training. You can work from anywhere, though being in Poland or Eastern Europe is a plus, and they want a quick start.