Project Based Recruitment | AI Quality Analyst - English | Remote

We need to output HTML fragment with sections using

for headings: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. Only include sections if posting has real specific content. We must not repeat job title as heading. No Job ID etc. Remove boilerplate, navigation, legal, EEO, duplicated lines, application instructions. Must preserve all substantive info. We need to parse the posting. Key info: - Project-based remote opportunity for AI Quality Analyst – English. - Three-month contractor project. - Focus: testing and evaluating a new AI personalization feature. - Role involves designing realistic multi-turn prompts, assessing how effectively an AI model uses permitted personal context, comparing model responses for grounding, integration, helpfulness, naturalness, and accuracy. - Suited to professionals with experience in AI evaluation, data annotation, content quality, linguistics, policy, ethics, journalism, research, or other analytical fields. - Candidates must be comfortable working in English, analyzing nuanced AI responses, and working independently in a remote environment. - Full-time availability in candidate's local time zone required because project operates as part of a global 24-hour team. - If exploring remote wfa jobs and have strong analytical and written communication skills, this project could be interesting. - Job Information: Position: AI Quality Analyst – English; Project Type: Project-Based Contractor; Work Arrangement: Remote; Engagement Length: 3 months; Commitment: At least 4 hours per day, up to 40 hours per week; Time Zone Requirement: 4 hours of overlap with PST; Language: English; Required Skill: Domain-Specific Languages; Equipment: Desktop or laptop with a reliable internet connection; Registration Link: REGISTRATION LINK HERE. - What Will You Do? (responsibilities): evaluate how effectively an AI system personalizes its responses using available user context. Create conversational scenarios based on realistic personal experiences and assess whether AI correctly understands and uses that context. Evaluation process involves 1–5 conversational turns. Examine whether personalization actually improves response. Consider whether AI uses relevant information appropriately, avoids unsupported assumptions, produces natural responses. - Key Responsibilities: list bullet points (we need to capture them). They include: Design creative multi-turn prompts based on personal context and experiences; Execute conversational scenarios designed to test AI personalization; Evaluate whether model responses correctly address the intent of the initial prompt; Check whether claims about the user are supported by available evidence; Identify incorrect personalization, hallucinations, and unsupported inferences; Assess how naturally personal information is integrated into responses; Identify unnecessary or excessive personalization; Compare two AI responses side-by-side; Determine which response is more helpful, natural, useful, and enjoyable; Write clear and defensible rationales for evaluation decisions; Reference specific conversation turns when explaining findings; Provide detailed annotations and constructive feedback; Review debug information to verify that relevant data sources and conversation summaries were properly utilized; Maintain data hygiene by removing evaluation conversations when required. - What Makes This AI Quality Analyst Role Different? (maybe part of responsibilities or description). It says: not conventional content-review; need to think from user's perspective and evaluator's perspective; determine whether personalization genuinely improves AI response or system making inaccurate/unnecessary/forced connections; need excellent judgment evaluating subtle differences. - Grounding and Personalization Evaluation: details about grounding: determine whether statements about user are supported by available info rather than assumptions/hallucinations. Identify unsupported assumption, misinterpret previous info, connect unrelated personal details, use outdated/irrelevant context, present inference as confirmed fact, miss important info. - Integration and Naturalness: assess integration; personal info should enhance response rather than make robotic; identify overnarrating; need attention to language, context, tone, user experience. - Side-by-Side Model Evaluation: compare two AI-generated responses side-by-side; goal determine which provides better overall experience; evaluation may consider helpfulness, accuracy, naturalness, personalization quality, relevance, ease of use, user intent, appropriate use of personal context, avoidance of unsupported claims, avoidance of unnecessary personalization; final ranking supported by concise defensible written rationale. - Required Qualifications: Candidates should demonstrate strong analytical and communication abilities. Important Qualifications Include: High-level English reading and writing ability; Strong analytical thinking; Ability to evaluate ambiguous and nuanced AI responses; Experience creating creative prompts; Ability to design multi-turn conversational scenarios; Understanding of personalization concepts; Strong attention to detail; Ability to compare AI responses and identify subtle differences; Excellent written communication skills; Ability to provide structured evaluation rationales; Ability to reference specific conversation turns; Ability to provide constructive feedback; Ability to work independently in a remote environment; Access to a desktop or laptop with reliable internet. - Education and Experience: A bachelor's degree or equivalent practical experience in a relevant field is preferred. Potentially Relevant Backgrounds Include: Policy, Law, Ethics, Linguistics, Journalism, Computer Science, Research, Data analysis, AI evaluation, Content quality. Previous experience in data annotation, AI quality evaluation, content moderation, or similar analytical work is strongly preferred. - Personal Account Requirement: use primary personal account rather than dedicated testing account; must be willing to use personal account and enable relevant personal data sources required for evaluation; because project evaluates personalized AI behavior, candidates should review privacy and data requirements before proceeding. - Working Commitment: At least 4 hours per day, up to 40 hours per week; 4 hours of overlap with PST; 3-month contractor engagement; Remote work environment; project operates as part of a global 24-hour operation, so flexibility within local time zone important. - Evaluation and Selection Process: steps: Job Interest Form, Profile Review, Assessment (must be completed within 24 hours), Pre-Onboarding. - Registration: link. - Why Consider This Project? etc. - Apply Early etc. - Conclusion etc. We need to produce HTML with sections: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. We need to decide which sections have real specific content. - About the role: we can summarize the role description (project-based remote opportunity, three-month contractor, evaluating AI personalization, etc). That's substantive. - Responsibilities: we have a list of responsibilities (design prompts, execute scenarios, evaluate, etc). That's substantive. - Requirements: we have requirements like full-time availability, 4 hours overlap with PST, English, desktop/laptop, remote work, personal account usage, etc. Also maybe "Ability to work independently in remote environment". That's requirements. - Qualifications: we have education/experience, preferred backgrounds, etc. That's qualifications. - Skills: we have specific skills like strong analytical thinking, English reading/writing, ability to evaluate ambiguous AI responses, experience creating prompts, design multi-turn scenarios, understanding personalization concepts, attention to detail, compare AI responses, written communication, structured rationales, reference conversation turns, constructive feedback, independent remote work, desktop/laptop. Could be considered skills. But we already have qualifications overlapping. The instruction says prefer a few substantive sections over many thin ones. We could combine Qualifications and Skills into one section? But they ask to use headings for those specific names if content exists. We need to decide which sections to include. We have Benefits: posting does not mention any benefits (like health, PTO). No mention. So we should not include Benefits. Pay: posting does not mention pay/salary. No mention. So we should not include Pay. Schedule: posting mentions commitment: at least 4 hours per day, up to 40 hours per week, 4 hours overlap with PST, 3-month engagement. That's schedule info. So we can include Schedule section. Thus sections: About the role, Responsibilities, Requirements, Qualifications, Skills, Schedule. Possibly also include "Requirements" for eligibility (availability, equipment, personal account). Qualifications for education/experience. Skills for abilities. We must not repeat job title as heading. So no

AI Quality Analyst – English

. Use headings for sections only. We must output only HTML fragment, no /, no markdown fences. We need to preserve all substantive information; do not invent or summarize away detail. So we need to include details like: three-month contractor project, evaluating AI personalization feature, designing realistic multi-turn prompts, assessing how effectively AI uses permitted personal context, comparing model responses for grounding, integration, helpfulness, naturalness, accuracy. Also mention suitability for professionals with experience in AI evaluation, data annotation, content quality, linguistics, policy, ethics, journalism, research, or other analytical fields. Must be comfortable working in English, analyzing nuanced AI responses, working independently remote. Full-time availability in local time zone required because project operates as part of global 24-hour team. If exploring remote wfa jobs and have strong analytical and written communication skills, this project could be interesting. We need to include the "Job Information" details: Position, Project Type, Work Arrangement, Engagement Length, Commitment, Time Zone Requirement, Language, Required Skill, Equipment, Registration Link. However we should not repeat job title as heading; but we can include those details in About the role or Requirements etc. The instruction: "Do NOT repeat the job title as a heading (the page already shows it) and do NOT put Job ID / requisition numbers as headings — fold minor metadata into a short

or drop it." So we can include the metadata in a paragraph. We must remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. So we need to strip out things like "Registration Link: REGISTRATION LINK HERE" repeated many times, also the "Why Consider This Project?" etc maybe considered application instructions? The instruction says remove application instructions. However we need to preserve substantive information; the registration link is likely application instruction. Should we drop it? The instruction says remove application instructions. So we should drop the registration link lines. Also drop "Apply Early", "Conclusion", "Why Consider This Project?" maybe considered promotional but still substantive? Might be considered boilerplate? The instruction: "Remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions." So we should remove sections that are purely about applying, like "Why Consider This Project?", "Apply Early", "Conclusion", "Registration Link". However we need to preserve substantive info about the role. The "Why Consider This Project?" includes some info about AI personalization importance, opportunity to work directly with AI quality evaluation, etc. That could be considered substantive but maybe it's promotional. Safer to keep it as part of About the role? The instruction says remove application instructions, not necessarily promotional content. But "Why Consider This Project?" is more like a call to action. Might be considered application instruction. I'd lean to remove it to be safe. Similarly "Apply Early" and "Conclusion" are application instructions. We also need to remove duplicated lines like repeated registration link, repeated bullet points. We need to keep the core description. Let's extract substantive content: - Role description: AI Quality Analyst – English, project-based remote, three-month contractor, evaluating new AI personalization feature, designing realistic multi-turn prompts, assessing how effectively AI model uses permitted personal context, comparing model responses for grounding, integration, helpfulness, naturalness, accuracy. - Suited to professionals with experience in AI evaluation, data annotation, content quality, linguistics, policy, ethics, journalism, research, or other analytical fields. - Candidates must be comfortable working in English, analyzing nuanced AI responses, working independently in remote environment. - Full-time availability in candidate's local time zone required because project operates as part of a global 24-hour team. - If exploring remote wfa jobs and have strong analytical and written communication skills, this project could be interesting. - Job Information (metadata): Position: AI Quality Analyst – English; Project Type: Project-Based Contractor; Work Arrangement: Remote; Engagement Length: 3 months; Commitment: At least 4 hours per day, up to 40 hours per week; Time Zone Requirement: 4 hours of overlap with PST; Language: English; Required Skill: Domain-Specific Languages; Equipment: Desktop or laptop with a reliable internet connection. - What Will You Do