Mercor is seeking Internet‑Native Bilingual Evaluator Experts (Spanish–English) to contribute to cutting‑edge AI research focused on multilingual communication, internet culture, and real‑world language use. In this role, you will evaluate, annotate, and improve AI model outputs across English and Spanish, helping ensure that models accurately understand modern online communication, cultural references, slang, and multilingual interactions. This role focuses specifically on Latin American Spanish internet culture, language use, and online communities across the region. Job Responsibilities Evaluate Internet‑Native Communication Analyse content originating from social media, forums, messaging platforms, online communities, and consumer‑facing digital environments. Assess AI‑generated responses for naturalness, fluency, cultural appropriateness, and internet authenticity. Identify issues involving slang, memes, abbreviations, code‑switching, and evolving online language. Evaluate content specific to Latin American internet culture,



trends, and regional online communities. Develop Evaluation Standards Create detailed rubrics and guidelines for evaluating bilingual and internet‑native communication. Define standards for tone, cultural relevance, contextual understanding, and linguistic accuracy. Document edge cases involving internet culture, humour, sarcasm, and platform‑specific language. Identify linguistic and cultural differences across major Spanish‑speaking countries in Latin America. Conduct Model Testing and Feedback Test language models using real‑world multilingual prompts. Evaluate outputs in both English and Spanish. Provide structured feedback to improve model performance and cultural understanding. Support Quality Assurance Participate in review cycles and benchmark development. Collaborate with project leads to improve evaluation frameworks. Main

📌 Bilingual Evaluator - Internet Culture Expert (Santiago)
🏢 Obsidian
📍 Santiago