Create product analytics challenges based on decisions you handle on a product team, then write reference solutions with clear reasoning. AI labs use these problems to assess model judgment. The work runs asynchronously, with no set schedule, and requires at least 10 hours each week. You can adjust your hours from week to week, with no stated maximum.
Apply your government or military SIGINT experience to projects that test AI systems. You will build unclassified scenarios, review generated intelligence assessments, and help set standards for sound analysis. The work is flexible and project- or task-based. Candidates complete an AI interview during the application process.
Use your hands-on experience with Premiere to work with video timelines and sequence assembly. You must have Premiere installed and ready to open on your own Windows PC. The role also requires screen captures above 2.5 megapixels, using one qualifying display or two monitors captured together. The project does not provide software licenses, virtual machines, or hardware.
Review AI-generated imaging responses or create expert content on modality choice, protocols, differential diagnoses, and radiology reports. You will assess clinical reasoning, accuracy, clarity, and safety, including CT, MRI, contrast, and radiation considerations. This remote, hourly contract has no immediate project; qualified experts may receive invitations when relevant work becomes available.
Bring at least three years of expertise from a professional field to annotation work for text, documents, model outputs, and other domain-specific material. You will assess content with informed judgment, write structured annotations, and support decisions with reasoning and citations. The role offers flexible scheduling, project experience for your portfolio, and exposure to AI development workflows.
Apply federal Employee and Labor Relations expertise to create and assess material for an AI evaluation project serving a government-focused customer. You will develop realistic cases, write policy-based reference answers, and review other experts’ work. No previous AI experience is needed; sound federal HR judgment and clear documentation matter most.
Take on real graphic design work on your own Mac to help a leading AI lab assess and improve its models. You’ll narrate your creative choices as you complete tasks that contribute to AI training data. The role calls for a Mac running macOS 15 or later, plus either current or recent master’s studies or the specified design education and experience.
Guide English quality across AI training projects by reviewing contributor work and generated content against project rubrics. You will spot-check submissions for accuracy, clarity, tone, grammar, formatting, and instruction-following, then share actionable feedback through Discord and maintain guidance in Google Docs and Google Sheets. The role also covers onboarding, contributor follow-ups, and consistent application of changing guidelines across remote teams.
Create and assess SQL and data modeling scenarios used to evaluate how frontier AI models handle professional analytics work. You will build examples from raw tables through models with tests, explain your modeling and query choices, and judge model answers for correctness and warehouse cost. The work is remote, async, and ongoing, with a minimum commitment of 10 hours per week and no fixed schedule.
Bring federal program and policy expertise to an AI evaluation project. You will write realistic agency scenarios, prepare reference materials, and assess work from other contributors against federal planning, evaluation, and reporting practices. The role draws on skills such as policy analysis, decision memo writing, and executive briefing preparation. No previous AI experience is required, but applicants must be based in the United States.
Record voice samples for a text-to-speech system that supports a client’s internal CX AI Agent. The client seeks women based in northern UK who speak with an authentic Northern English accent, including Yorkshire, Manchester, Newcastle, or Liverpool varieties. The work involves one recording session of about four hours, with a possible second session of similar length. The client will not sell, license, or reuse recordings for another purpose.
Review AI-generated Lithuanian and write reference answers that show accurate, natural language use. You will compare model responses, assess grammar, meaning, tone, and cultural fit, then explain issues and suggest corrections in English. The role requires native or near-native Lithuanian, C1 English, and at least three years of language-focused experience; the work follows an hourly remote contractor arrangement.
AfterQuery is assembling senior experts in ML inference and GPU kernel engineering to assess how frontier AI models handle demanding technical problems. You will shape evaluation tasks and judge model responses using your own domain expertise. The work centers on research and evaluation, not production engineering, and covers inference serving, kernel optimization, and large-model deployment infrastructure.
Contribute federal audit expertise to an AI evaluation project by creating and reviewing realistic audit scenarios and reference answers. You’ll check that project materials reflect federal oversight practices and real audit processes. No prior AI experience is needed; a background in performance or financial auditing is the focus.
Bring your medical coding and documentation expertise to the review of clinical records and AI-generated notes. You will check coding accuracy, annotate documentation, and share feedback that helps improve clinical AI tools. Applicants must live in the US, hold an active qualifying certification, and speak, read, and write English plus another language at C1 level or above.
Review Bahasa Melayu AI responses or create reference answers across general knowledge, customer support, education, and conversation. You will judge grammar, meaning, tone, factual claims, and cultural fit, then explain issues in written English. This remote contract has no immediate project; qualified experts may hear about relevant work through the expert network.
AfterQuery offers project-based work for supply chain professionals with 1 to 3 years of experience. Over the next 2 to 3 weeks, you’ll review plans and operational documents, assess forecasts and inventory strategies, and examine sourcing and distribution decisions. Projects run asynchronously, with a commitment of 10 to 20 hours per week for work you accept.
As a Federal Regulatory Policy Analyst, you will create realistic rulemaking assignments and assess public comments for a project supporting a federal government customer. You will also draft agency responses and review other contributors’ work. Your expertise in federal rulemaking and regulatory policy will help train AI systems; prior AI experience is not required.
Use your legal practice experience to help a leading AI lab assess and improve how its models handle real legal work. You will review legal tasks and model responses, write task instructions and reference solutions, and create evaluations alongside research teams. The role requires 8 to 15 years of post-qualification practice, U.S. bar admission, and a Bay Area residence for a hybrid schedule. Cincinnatus LLC employs this full-time W-2 role for an initial six-month engagement.
As an Anthropology Quality Assurance Lead, you will review AI-generated anthropology content and contributor work against project rubrics, then give clear feedback to trainers and QAs. You will also maintain review materials and help contributors apply consistent standards across remote projects. This is a contract role with no immediate project; qualified experts may be contacted for future work. Selection includes an AI interview, a domain task, and a recruiter interview.
Location and eligibility
Remote · United States only
Engagement type
Contract
Compensation
Up to $65/hr
anthropology
cultural anthropology
archaeology
Quality Assurance & TestingView job
Showing 20 active roles of 1,907. 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.