Record first-person videos at home with a compatible smartphone while completing specified physical activities. The footage and phone IMU data support AI training. You do not need prior AI experience, but you must follow detailed collection protocols and submit 10 to 15 hours of video each week, including at least 10 hours of approved footage. Payment is based on each video that meets project specifications.
Review AI coding attempts and determine whether failures stem from the code, benchmark tests, task instructions, or execution environment. You’ll examine Python and other code, test behavior, and run logs, then document evidence-based conclusions. The contract requires 35 hours per week and is open to candidates in the United States. Work starts immediately.
Take part in a paid study of a new desktop AI assistant. For at least four weeks, install the tool on your primary work computer and use it during your usual workday. It records your screen as you work, so you must be comfortable sharing your active work sessions and keeping a schedule of about 40 hours per week. You may also provide brief feedback at the end.
Audit complete AI-agent shopping and delivery tasks, including the prompt and grading rubric. You will judge whether each rubric failure was fair by checking the available information, the prompt’s requirements, the treatment of valid answers, and the rubric’s structure. You will also give reviewers specific feedback on prompt and rubric issues.
Use your mechanical design experience to create realistic engineering scenarios and assess AI-generated work. You’ll judge technical accuracy, manufacturability, and alignment with real CAD, tolerancing, and manufacturing workflows, then write structured feedback. The project-based work has no set schedule or minimum hour requirement and can fit alongside your current role.
Review AI-generated answers in Catalan and English, judging language quality, factual accuracy, and how well each response follows its prompt. You will score and compare model outputs, write prompts and model explanations, and record decisions in English. The work calls for advanced Catalan proficiency, structured linguistic QA, and careful use of reliable references.
AfterQuery hires experienced Scrum Masters as independent contractors to create and evaluate scenarios that test how AI models handle agile delivery. You’ll develop examples spanning sprint planning through retrospectives, explain sound facilitation choices, and assess model responses for their effect on real teams. The work runs remotely and asynchronously, with a minimum commitment of 10 hours per week and ongoing project assignments.
Help a foundational AI lab train its language model on professional presentation work. Create, assess, and refine AI-generated decks across core product marketing workflows. Work remotely on your own schedule, with hours tied to client task volume. Most experts work under 10 hours per week, though no weekly hours cap applies. The start date is ASAP.
Reviewers assess AI-generated content for accuracy, clarity, natural phrasing, and fit with project guidelines. You need C1-level fluency or higher in one of the requested languages and a strong grasp of its meaning, tone, phrasing, and cultural context. Projects come in batches, and availability depends on the language and project needs.
Review customer-facing shopping and product discovery experiences, simulate concierge support, and improve conversational content. The work supports training AI systems with real-world input, with a focus on Shopping QA, Concierge Services, and Product Discovery.
Assess AI responses in behavioral health by creating clinical scenarios and judging therapeutic fit, empathy, safety, and clinical reasoning. The six-week remote contract is for US-based clinicians and starts immediately. It requires eight hours of work per day, including four hours of overlap with Pacific Time. The project starts with paid certification hours; clinicians who pass can move into production work.
Take part in a paid, remote 20-minute voice or video interview about vehicle ownership and connected car services. You’ll discuss your commute and use of built-in features, compare those tools with mobile alternatives, and react to concepts for in-car digital services. The study seeks feedback to help refine automotive software and interfaces. Applicants must be in the USA.
Contribute systems-level expertise to AI training work for a leading AI lab. You’ll create and assess technical tasks on embedded systems, kernels, firmware, device drivers, and hardware design, then write feedback and evaluation rubrics. Cincinnatus LLC employs this part-time, remote role as a W-2 position and places you with the client team. The schedule requires at least 20 hours each weekday week.
The year-round Handshake AI Fellowship brings engineering specialists into project-based work that helps improve large language models in expert fields. You’ll create subject-specific prompts, assess model responses, and research topics with AI. Work remotely on your own schedule for about 10 to 20 hours each week; project placement depends on availability.
Lead quality reviews for Traditional Chinese Hong Kong and Cantonese AI training projects. You will assess generated content and contributor work against project rubrics, give actionable feedback in English, and guide remote trainers and reviewers through Hong Kong-specific language standards. The role is a contract opportunity with no immediate project; qualified experts may be contacted when relevant work becomes available.
Create and assess office management challenges used to train AI models. You will draw on experience in administrative leadership to develop scenarios about team supervision, operational workflows, and practical problem-solving, then check model answers against established office management standards. The role is remote and asynchronous, with a flexible schedule.
Help train a language model by creating, reviewing, and improving professional presentations for core market access and HEOR workflows. You’ll work remotely on your own schedule, with hours tied to client task volume and not guaranteed. Most experts work under 10 hours per week, with no weekly cap on hours or earnings. The expected start is ASAP.
Review AI-generated coding and technical customer service content, including virtual agent replies, JSON, and audio-related material. You’ll check accuracy, clarity, natural language, and usefulness against project guidelines. The role requires native-level Brazilian Portuguese, but you do not need to live in Brazil. Selection includes a language introduction and a JSON assessment.
Location and eligibility
Remote · Eligibility not specified
Engagement type
Not specified
Compensation
$11/hr
brazilian portuguese
json
coding
Language & LinguisticsView 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.