Create reference advertising edits and assess AI-generated video for training and evaluation projects. You’ll bring studio or agency editing experience to hands-on post-production, build pacing and continuity standards, and give specific feedback on model output. The work covers commercial and social content in several aspect ratios, with a 15-minute AI interview for shortlisted candidates and a 45-minute technical interview for those who advance.
Review GPU and accelerator kernel-development tasks that support training and evaluation of a frontier AI lab’s models. Judge numerical correctness, benchmark fairness, task scope, and compilation and runtime validity across varied kernel work. You’ll also write rubric-based feedback that explains your findings.
Use your energy industry experience to create expert training material and assess AI-generated responses about utility metering, customer operations, and billing systems. You’ll check responses for accuracy and relevance, then share feedback on utility workflows and customer data practices. The remote, project-based work has flexible hours, no minimum commitment, and no AI experience requirement.
Review AI-generated medical responses or create healthcare content that demonstrates sound clinical reasoning. You will check evidence, test interpretation, timelines, dosing logic, and care-setting choices, then explain errors and compare answers with clear written feedback. This remote, hourly contract has no immediate project; qualified experts may hear about future work through the network.
Help train and evaluate AI image models by supplying original images and comparing them with generated results. You’ll identify visual flaws, assess composition and fidelity, and share clear observations through written and verbal channels. The role is for candidates based in the United States, with experience in visual arts or image evaluation considered useful.
Create professional audio examples and reference material for AI training and evaluation. The work covers rights-cleared, multi-stem music, commercial video mixes, voiceover cleanup, Foley, and sound effects, along with loudness and synchronization checks. The contract expects 39 hours per week and requires a United States base, a live portfolio, and participation in a 15-minute AI interview followed by a 45-minute technical interview for candidates who advance.
Review repository-level software engineering benchmarks used to train and evaluate a frontier AI lab’s models. You’ll examine task quality, reference patches, test harnesses, Docker isolation, and grading integrity, then write clear feedback against a rubric. The role calls for professional software engineering experience and open-source contribution or maintainer work. The listing is limited to the United States.
Use your power engineering experience to create realistic technical scenarios and assess AI-generated answers. You’ll check engineering calculations and recommendations for accuracy, safety, practical feasibility, and consistency with industry practice. The work covers areas such as grid operations, microgrids, and protection and control. Set your own pace, with no fixed schedule or minimum-hour requirement.
Review AI-generated English responses or create language examples that show sound reasoning and clear, correct writing. You will judge accuracy, prompt fit, and meaning, then explain edits or compare candidate responses. The work draws on English grammar, discourse analysis, and tools such as ChatGPT, Gemini, and Perplexity. This remote contract role requires at least C1 English proficiency and a relevant bachelor’s degree.
As a Digital Designer, you’ll contribute original images or photographs and compare them with AI-generated versions. You’ll assess visual differences, identify image defects, and write precise feedback that helps improve AI image systems. No previous AI experience is required. The work is remote and limited to people located in the United States.
Take part in an in-person study in Midtown Manhattan about how adults use smartphones, apps, and digital tools in different languages. You’ll demonstrate your everyday device habits, share examples within the study’s consent and privacy rules, and complete a short questionnaire. The study requires 16 hours across two consecutive days, with a possible third day for up to 24 hours. Sessions run through the end of November. Participants who complete the study receive $1,200–$1,600 under the terms provided before enrollment.
Apply your outpatient primary care experience to review clinical notes and assess AI-generated documentation for accuracy, completeness, and clinical soundness. You will annotate encounter data, flag gaps or errors, and give structured feedback that helps improve clinical documentation and decision-support tools. Applicants must practice in the US, hold an MD, and speak, listen, and write in English plus another listed language at C1 level or higher.
Licensed electricians and other electrical professionals can contribute field expertise to AI research through remote, project-based contract work. You’ll create expert training data and assess AI responses about electrical work, drawing on experience with installation, wiring, troubleshooting, and code compliance. The schedule is flexible, with no minimum hours; contributors often work 5–20 hours per week while a project is active.
Use your mathematics education and Gujarati expertise to shape AI training content. You will review generated responses, write reference answers, and compare outputs for accuracy, grammar, tone, and cultural fit. The work spans prompts in areas such as education, customer support, general knowledge, and conversational writing, with written feedback in English.
Work on a U.S. Navy project that uses drone and maritime sensor imagery to train AI models. You’ll review machine-generated bounding boxes, add objects the models miss, and follow project guidelines in a secure Navy environment. The project may expand from still images to video frames. Compensation is based on each completed image submission, and prior AI experience is not required.
Create evaluation tasks that ask a model to design and refine passive on-chip RF components against electrical and area targets. For each task, provide the specification, a reference layout, and a check that runs an EM simulation and compares its results with the target. The work starts with five pilot tasks, followed by a review before the project scales. Applicants must be based in Europe, Canada, or the US.
Lead AI training data, human data, or model evaluation programs for a leading AI lab. You’ll coordinate domain experts, turn research and product input into clear task guidance, and track quality and throughput. The role requires 40 hours per week on weekdays. Cincinnatus LLC employs the selected candidate as a W-2 employee and places them with the client’s team.
Use your critical care training and ICU experience to shape AI research. You’ll create challenging clinical tasks, assess AI responses for accuracy and patient safety, and write feedback grounded in current practice and guidelines. The work is remote, project-based, and part-time, with no set schedule or minimum hours. Projects open throughout the year, so you can contribute alongside your current role.
Guide quality checks for physics AI training by reviewing generated explanations, calculations, diagrams, and contributor work against project rubrics. You will give specific feedback, answer physics and workflow questions, maintain contributor resources, and address recurring quality gaps. The role is remote and contract-based, requires native Punjabi fluency, and includes an AI interview, a domain task, and a recruiter interview.
Prepare and assess TY2025 Form 1040 returns as part of a project that uses expert tax work to train AI systems. You’ll build returns from the facts provided, document how each figure was calculated, create complex tax scenarios, and review AI-generated computations. The role is remote and limited to applicants in the US. No AI experience is required.
Showing 20 active roles of 683. 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.