Remote AI training and data labeling jobs

Project role comparing outputs from several chatbots against business requirements and writing a final assessment per task. 16 weeks, five days a week including weekends as needed. Requires 2+ years as a business analyst. No rate published.

Review and rate the actions and answers of a computer-using AI agent, verify claims by online research and write detailed annotations. Hourly long-term engagement, 16 weeks with possible extension; no rate published. 45 to 50 minute form assessment.

Hourly, 16-week engagement on a computer-using-agent evaluation project: review AI responses and actions, verify claims online and write ratings and explanations. A 45 to 50 minute Google Form assessment comes first. No rate published.

Computational biologists with a PhD or equivalent design agentic life-science tasks where AI must navigate data, run analyses and troubleshoot to a valid result. $300 per approved task, 40+ hours a week, 5-week contract, 4-hour Pacific Time overlap.

Mathematicians and statisticians with a PhD or equivalent build terminal tasks (optimisation, ODEs, stochastic modelling, numerical methods) with rigorous graders for Terminal Bench Science. $300 per approved task, 40+ hours a week, 5-week contract.

Earth scientists with a PhD or equivalent build terminal-based benchmark tasks (geospatial, climate, ocean, geology) that test AI agents on Terminal Bench Science. $300 per approved task, 40+ hours a week, 5-week contract, 4-hour overlap with Pacific Time.

Engineers with a PhD or equivalent design multi-step terminal tasks from real engineering workflows (simulation, optimisation, control, signal processing) to train and evaluate AI agents. $300 per approved task, 40+ hours a week, 5-week contract.

Vietnamese speakers with a master's degree and 3+ years of work experience write prompts and evaluate AI answers for Turing. Native or professional Vietnamese plus strong English. Assessment before start.

Computational life scientists design multi-step tasks that AI agents must solve by writing and running code on scientific data. Paid $280 per approved task, 5-week contract, 40+ hours a week with 4 hours of PST overlap.

PhD-level bioinformaticians design hard agentic genomics tasks, datasets, reference solutions and graders for GeneBench Pro. Contract, 5 weeks, at least 40 hours a week with PST overlap. $260 per approved task.

Korean-speaking photographers source licensed real-world images, obscure key details and write English prompts, rubrics and solution notes so multimodal agents must research to identify them. Contractor, 20+ hours a week, up to 16 weeks. No rate published.

Experienced engineers review what AI coding agents produce in real repositories, build rubrics and label preference data. $100-150/hour, 20-40 hours a week, about 3 months, open to North America only, with at least 6 hours of Pacific Time overlap.

Contract evaluator for a Gemini personalization feature: write prompts from your own life, judge whether the model used your Gmail, Search and YouTube history well, and rank side-by-side answers. 4 to 40 hours a week, 3 months, 4 hours of PST overlap. No rate published.

Turing contractors build a benchmark for AI browsing agents by writing research questions the models cannot solve, each backed by an auditable evidence trail. About $30 per approved task, 8 weeks, up to 40 hours a week, fully remote.

Students and professionals with real Gemini learning history research, verify and correct AI responses. $30 per hour, 16 weeks, 20 hours a week (up to 40), paid in USD. Your Gemini conversation history is validated.

Sports journalists and analysts probe AI models on leagues, athletes, rules and statistics, writing prompts, ranking answers and building benchmarks. 8-week contractor assignment, 40 hours a week. No rate published.

Policy, government and political research professionals evaluate AI answers on elections, governance, public policy and international relations, and build adversarial test sets. 8-week contractor assignment, 40 hours a week. No rate published.

Musicians, producers and musicologists test AI answers on theory, genres, composers, instruments and production, writing prompts and benchmark cases. 8-week contractor assignment, 40 hours a week. No rate published.

Art historians, curators and design professionals write adversarial prompts and grade AI answers on fine arts, architecture, museums and artists. 8-week contractor assignment, 40 hours a week. No rate published.

Tabletop and strategy-game experts build and grade rule-based reasoning tasks for AI models. Contractor assignment of 2 months, 20+ hours a week with PST overlap, take-home assessment. No rate published.

Small business owners write realistic prompts, chat with several AI chatbots (up to 5 turns each) and compare the answers for clarity, usefulness and accuracy. A 16 week project-based contract; you need business documents in English. No rate published.

Evaluate Gemini's personalization in Portuguese using your own Google account data, ranking side-by-side responses and writing rationales. $15 per hour, 3 months, 4 to 40 hours a week with 4 hours of PST overlap.

Business owners who can supply Japanese business documents evaluate and compare AI chatbot replies to small business scenarios, then submit transcripts and assessments. Project-based, 16 weeks, remote contract; no rate published.

Small business owners with Korean business documents compare AI chatbot answers on practical scenarios and write structured evaluations. Project-based, 10 weeks (the shortest of three language variants), remote contract; no rate published.