AI evaluation roles ask you to judge how well a model performed. That can mean scoring responses against a rubric, identifying factual errors, comparing two answers, or testing a model against professional standards.
Review candidate CVs and recorded AI interviews to support hiring decisions across technical and non-technical roles. You will assess qualifications, identify relevant experience, and write structured recommendations that help maintain fair, consistent hiring evaluations. The role requires fluent English, a bachelor’s degree or higher, and a personal computer suitable for reviewing candidate materials and completing evaluation work.
Use your epidemiology and public health expertise to assess AI-generated answers or write reference prompts and model responses. You will check population-health reasoning, evidence interpretation, surveillance logic, and recommendations for accuracy, clarity, and safety. This remote, hourly contract requires at least five years of professional public health experience and C1 English proficiency; qualified experts may be contacted when relevant projects arise.
Use your teaching expertise to assess educational materials and help develop training data for AI systems. You’ll review AI-generated lessons, quizzes, rubrics, student feedback, and parent communications, then flag accuracy, age-appropriateness, and teaching-practice concerns. No AI background is needed, but you must work full time in education and have at least two years of hands-on teaching experience.
Bring hands-on capital markets compliance expertise to the evaluation of AI systems that detect suspicious trading and potential market abuse. You’ll review AI-generated alerts and investigations, assess trading patterns and electronic communications, and provide structured feedback on the accuracy of investigative findings and surveillance logic. Applicants must be based in the United States and have at least four years of relevant professional experience.
Contribute counterintelligence expertise to the development and evaluation of AI systems for security applications. You’ll assess case studies and model responses against real-world practice, legal authorities, and oversight rules, with particular attention to the line between valid threat analysis and prohibited surveillance or targeting. Prior AI experience is not required; counterintelligence knowledge is the focus.
Use your public-sector expertise to create and assess tasks that help train AI models on real government workflows. You will write, review, and grade policy prompts covering public procurement, budget planning, fiscal analysis, and program evaluation through a desktop platform. The project is remote and requires at least 20 hours per week through December.
Audit AI agents’ work on realistic shopping and delivery tasks by reviewing the prompt, grading rubric, and full agent transcript. You’ll judge whether the rubric treats failures fairly, follows the guidelines, and matches the prompt, then give reviewers specific feedback on prompt and rubric issues. Applicants need substantial experience writing or reviewing rubrics for LLM evaluations and must list those projects.
Take part in a paid, video-based research interview of about 45 minutes on how organizations select post-training platforms and services. Discuss your use case, provider choices, evaluation process, data and evaluation work, costs, results, and business value. Interviews can be scheduled at your convenience, and the client will not request confidential code or proprietary data.
Use your applied statistics, biostatistics, actuarial science, or related expertise to create research-level problems that test and train AI models. AfterQuery works with leading AI labs and seeks contributors who know real operational, survey, or insurance data. The remote contract pays $300 per approved task, with each task taking about six hours. You can choose your workload, with no fixed weekly schedule or cap on the number of tasks.
Use your retail pricing experience on an ongoing, project-based AI research opportunity. You’ll create training scenarios, assess AI responses against real pricing and markdown practices, and write feedback to guide model improvement. The part-time work is remote and self-scheduled, with no set hours or minimum hour requirement, so you can take it on alongside your current work.
As an independent contractor on an audio and language model evaluation project, you’ll assess Dutch samples for accuracy, fluency, naturalness, and overall quality. Your reviews may require close attention to grammar, pronunciation, tone, phrasing, and word choice, including whether spoken Dutch fits its context. Each task takes about 30 seconds to 4 minutes.
Evaluators review model outputs for correctness, relevance, completeness, and safety. Many projects use structured rubrics, so the work rewards careful reading and consistent judgment more than speed alone.
Subject knowledge matters. A finance evaluation project may require professional finance experience; a medical evaluation project may require a clinical background.
Skills that help
Clear written feedback is central to evaluation work. You should be able to explain why an answer is wrong, not only mark it as wrong.
Familiarity with the domain, attention to detail, and comfort working through structured guidelines are usually more relevant than prior AI experience.