This description is a summary of our understanding of the job description. Click on 'Apply' button to find out more. Role Description Mercor is seeking native Simplified Chinese speakers with exceptional writing skills to contribute to a high-impact AI research project with a leading lab. Freelancers will author Simplified Chinese/English prompt–golden answer pairs that train and evaluate advanced language models. This is a short-term, flexible opportunity for professionals who combine language mastery, strong critical thinking, and a knack for instructional clarity. Ideal for those who enjoy distilling complex concepts into well-crafted, culturally grounded Simplified Chinese text while maintaining technical precision in English. Job Details - Multilingual Prompt Design & Optimization: Create detailed prompts in Simplified Chinese and/or English with multiple constraints and instructions, ensuring natural phrasing and real-world relevance for Simplified Chinese–speaking users. - Define and Document Evaluation Standards: Establish high-level expectations for correct responses in Simplified Chinese consumer contexts, and develop comprehensive rubrics that account for linguistic nuance, tone, and cultural conventions. - Model Testing and Grading (Bilingual): Run prompts through models and assess preliminary outputs for accuracy, fluency, and cultural fit in Simplified Chinese, comparing results against English where needed. - Benchmarking & Quality Assurance: Collaborate in QA review processes to ensure prompt tasks and rubrics meet rigor—maintaining consistency and reliability across Simplified Chinese–language benchmarks before integration into official evaluations. Qualifications - Native-level fluency in Simplified Chinese (written) with strong reading/writing ability in English. - BS or BA from a reputable institution (completed or in progress). - Strong writing and critical thinking skills. - Ability to work independently and meet deadlines. - Significant familiarity with ChatGPT or similar tools for personal decision-making, hobbies, or general interests. - Able to reliably produce China–specific, culturally accurate Simplified Chinese. Preferred Qualifications - Experience in teaching, research, editing, or academic writing. - Experience creating evaluation criteria, rubrics, or grading guidelines. - Familiarity with LLMs, prompting, or model evaluation (helpful but not required). Application & Onboarding Process - Complete an AI-led interview (about 15 minutes). - Complete a 45-minute written assessment focused on writing and rubric creation. - If selected, you will be invited to work on the project. More Details About This Role - Expect to contribute at least 20 hours per week. - Expect a commitment of around 2+ months. - You’ll be working in a structured project environment with clear goals and tools.

• Drive forward-looking ML research to shape Nebius’ AI platform and PaaS roadmap, translating frontier developments into clear product direction and priorities. • Convert state-of-the-art ML pipeline insights into actionable requirements, reference architectures, benchmarks, and gap analyses. • Partner cross-functionally with Product, Engineering, and ML teams to align platform capabilities with emerging ML workloads and best-practice stacks. • Build strategic collaborations with universities, research labs, and the broader ML ecosystem to accelerate innovation and credibility. • Establish quality standards for ML-enabled services, including evaluation rigor, reproducibility, reliability, and responsible ML practices. • Engage strategic customers to understand complex ML scenarios and translate them into clear functional and non-functional requirements. • Provide senior technical leadership during evaluations, architecture reviews, and escalations, ensuring customer realities inform platform decisions. • Articulate and communicate a clear vision for AI-enabled applications and the infrastructure stack required to support them, influencing both technical and executive audiences.

• Construct expert benchmarks: Build and validate real-world investment cases, securities analyses, and portfolio management frameworks used to evaluate frontier AI systems. • Stress-test model reasoning: Diagnose weaknesses in AI-generated investment analyses, identifying where logic, assumptions, or market intuition fail. • Design frameworks: Translate how institutional investors evaluate securities, construct portfolios, and manage risk into problems that push the limits of AI reasoning. • Compare human vs. machine judgment: Systematically evaluate divergence between professional investment reasoning and AI outputs across equities, fixed income, and multi-asset contexts.

• Construct expert benchmarks by building and validating real-world underwriting cases, actuarial analyses, and investment frameworks used to evaluate frontier AI systems. • Stress-test model reasoning by diagnosing weaknesses in AI-generated insurance analyses, identifying where logic, assumptions, or domain intuition fail. • Design frameworks that translate how insurance professionals evaluate risk, price policies, manage general account portfolios, and model liabilities into problems that push the limits of AI reasoning. • Compare human vs. machine judgment by systematically evaluating divergence between professional insurance judgment and AI outputs across investments, underwriting, and actuarial contexts.