The work
You will evaluate AI reasoning on complex technical problems, using your specialist knowledge to define what correct, complete work looks like. This is evaluation and research work, not production engineering.
- Design realistic technical scenarios and problem sets in your area of expertise.
- Write expert reference solutions and grading rubrics.
- Review AI-generated answers for correctness, depth, and sound domain judgment.
- Write clear technical explanations and structured feedback.
What it pays and takes
- Pay: $135 per hour.
- Time: 20+ hours per week; flexible and asynchronous.
- Location: Remote, open to candidates in the eligible countries listed for this role.
- Language: Professional fluency in English and strong written communication.
- You must have current or recent hands-on professional activity in at least one relevant specialist field and meet the experience and education requirements for at least one track.
- You must be able to work independently in a remote setting.
- Relevant fields include software or systems engineering, formal methods, computational science, machine learning inference, GPU kernels, enterprise operations, security, hardware design, and creative technology.
- Experience with technical review, grading, code review, peer review, editing, research publications, or recognized work products is helpful. Track expertise may include theorem proving, reverse engineering, cryptanalysis, RTL, CAD, robotics, audio engineering, linguistics, or multimodal design.
- This is part-time contract work.
How it works
Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI training work
AI training is the human work behind building AI systems: people prepare examples and assess model answers. Specialists contribute their expertise to help judge whether those answers are accurate and well reasoned.
Listed by OpenTrain AI. Details can change, so check the original posting before applying. View original posting