Review AI-generated JavaScript code and explanations, or write model answers that show sound methods and clear reasoning. You will judge correctness, readability, performance, and prompt adherence, then explain logic errors and edge cases in strong written English. The work draws on modern JavaScript, TypeScript, and experience with a frontend framework or backend runtime. No project is active now; qualified experts may hear about future assignments.
A PhD-level scientist will create and assess expert scientific material used to train AI systems. The work centers on microbiome and metagenomics, clinical and translational bioinformatics, and related life sciences, with an emphasis on sound evidence, verifiable answers, and clear scientific reasoning.
Create and assess HR questions and case studies that help train AI models to handle real workplace scenarios. You will test model responses against HR practices and employment regulations, then work with HR and technical specialists to shape content that challenges models on recruitment, employee relations, compensation, and other core areas.
Use your clinical judgment to help AI research teams assess model responses. Work may include creating clinical vignettes, grading diagnostic reasoning, and identifying recommendations that fail to account for a patient’s circumstances. This standing physician pool is not a specific opening; Mercor contacts candidates when a project matches their background, which may take a week or several months. Apply with a resume, complete an AI interview of about 20 minutes, and verify your license and background.
Use your mathematics research experience on flexible, project-based work that helps AI models handle mathematical reasoning. You will create expert-level problems, assess model responses for accuracy and rigor, and write feedback grounded in real proof, formalization, and computational practice. Work remotely at your own pace, with no set schedule or minimum hours.
Review AI-generated answers and create Korean-language content that demonstrates sound reasoning, accurate facts, and clear explanations. You will apply South Korean localization judgment to assess terminology, tone, cultural nuance, and prompt fit, using MQM/LQA concepts to classify quality issues. This remote contract role considers candidates based in South Korea who have at least five years of localization or translation experience.
Create expert-level life sciences problems and assess AI responses for scientific accuracy, rigor, and depth of reasoning. You’ll draw on research experience to build verifiable answers and evaluation rubrics, using primary literature and established experimental or computational methods. The work calls for sound judgment across both wet-lab and computational questions, plus written explanations that other experts can review.
Create and assess presentation design challenges that help train AI models to handle real-world slide work. You’ll draw on PowerPoint expertise and visual communication principles to build scenarios around layouts, charts, and storytelling, then judge how well model responses apply those principles. The role is remote and asynchronous, with a flexible schedule.
Join a standing pool for project-based AI training and evaluation work across the creative fields. You may write briefs and grading references based on completed work, assess model output in areas such as graphic design, UX/UI, writing, and imagery, and explain when an answer misses the standards of its medium. After a background check and a roughly 20-minute AI interview, Mercor may contact you when a project fits your discipline. Timing can range from a week to several months.
Use your AEC and Vectorworks experience on an ongoing, project-based AI research effort. You’ll create realistic BIM tasks, assess AI responses, and write structured feedback tied to workflows such as modeling, construction documents, and project coordination. The work is remote, self-scheduled, and has no set schedule or minimum hour requirement, so you can fit it alongside your current role.
Build realistic ClickUp workflows that show how teams manage projects, operations, and collaboration. You will prepare workspace data, carry out tasks such as configuring views or automations, then record your screen and document the work in SuperAnnotate. Your submissions will support AI training by showing models how people use project-management software in practical scenarios.
A PhD-level biology or bioinformatics expert will assess workflows and biological data for AI benchmarking and model development. You’ll review analysis methods across areas such as genomics, transcriptomics, proteomics, and computational biology, then share recommendations and document your findings. Applicants need hands-on experience with real biological data and strong Python or R skills.
AfterQuery creates evaluation data that AI labs use to assess and improve their models. You’ll draw on your own research to develop and solve machine learning analysis problems covering experiment design, ablations, model evaluation, and behavior. The role suits hands-on researchers rather than engineers focused on deployment, data engineers, or annotators. Work remotely with an asynchronous schedule and no fixed hours.
Mercor keeps this listing open to build a pool of machine learning practitioners for AI training and evaluation projects. You may write problems from systems you have shipped, assess model-generated code, or spot approaches that could fail in production. Apply with your resume, confirm where you work from, and complete an AI interview that takes about 20 minutes. Project invitations can arrive within a week or several months.
Use your energy operations experience to create training examples and assess AI-generated material about drilling workflows. The remote, project-based contract work offers a flexible, asynchronous schedule with no minimum hours. Contributors often work about 5–20 hours per week during active projects. Placement depends on project needs, and future projects may become available. No AI experience is required.
Lead Dutch quality work across AI training projects by reviewing generated content and contributor submissions against project rubrics. You will assess Dutch language quality across grammar, tone, cultural fit, and meaning, then guide trainers and QAs through feedback, onboarding, and project updates. The role also calls for clear English communication and organized upkeep of quality documentation for remote teams.
Review legal scenarios and AI-generated legal work for a project aimed at improving legal reasoning and accuracy in AI systems. You will assess liability, spot gaps or errors, and give structured corrections grounded in your legal expertise. Applicants must hold a J.D. from a U.S. law school, maintain an active license in at least one U.S. state, and live in the United States.
Location and eligibility
Remote · United States only
Engagement type
Contract
Compensation
$100–$170/hr
legal reasoning
issue-spotting
document review
LegalView job
Showing 20 active roles of 1,328. Dates reflect what the source published, not when AIWorkJobs last synced.
Sources: ten specialist AI work platforms. AIWorkJobs does not verify every claim made by a source. Report a listing that looks wrong through the contact page.