Evaluate AI-generated answers to real Japanese law questions by comparing two LLM outputs, scoring them against a rubric, and providing concise feedback. This remote contract pays $140–$150 per hour.
Legal & Finance
Remote Hourly · $140–$150/hr
$140–$150/hr
Compensation
1 country
Eligibility
Entry
Experience
Aug 16, 2026
Posted
Open to applicants in
Japan
Interested in this role?
Create a free OpenTrain account and apply in minutes.
OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. OpenTrain AI hires and contracts contributors for projects that help shape how modern AI systems understand, reason about, and respond to the world.
About AI Evaluation Work
AI evaluation is the human side of building reliable artificial intelligence. Subject-matter experts review model outputs, identify strengths and weaknesses, and provide structured feedback that helps improve how AI responds to real-world questions.
Work remotely with flexible project-based opportunities.
Apply your professional expertise to cutting-edge AI systems.
Help evaluate and improve the quality of legal AI responses.
The Role
OpenTrain AI is seeking an expert Japanese law evaluator to benchmark AI responses to real questions of Japanese law. You will run prompts through two large language model platforms, compare the outputs, score each response using a standardized rubric, and provide concise written feedback.
This is a short-term, high-impact contract with meaningful potential for follow-on work. The initial commitment is approximately 10 hours. The listed schedule indicates availability of 20+ hours per week.
Contractor, part-time engagement
Pay: $140–$150 USD per hour
Location: Japan
Language: Japanese
Data type: Text
Work type: Evaluation and rating
What You'll Do
You will independently assess the legal quality and comparative performance of AI-generated responses while following the project rubric and meeting deadlines.
Run the same set of prompts across two AI platforms.
Compare both platforms' outputs side by side.
Score each response against a standardized rubric.
Submit concise written feedback for every evaluation.
Requirements
This role requires professional legal experience and qualification to practise in Japan. Strong knowledge of Japanese litigation and procedural law is essential, along with precise written communication in Japanese.
At least 2 years of law-firm experience in litigation, disputes, or contentious practice; 3 or more years preferred.
Qualified to practise in Japan as a bengoshi or gaikokuho jimu bengoshi.
Strong command of Japanese litigation and procedural law.
Native or fluent Japanese with precise written communication.
Ability to work independently according to a rubric and deadlines.
Why Work in AI Training
Every major AI system depends on examples and expert judgments prepared by people. By evaluating legal responses, you can contribute directly to the development of more capable and reliable AI while applying your professional knowledge in flexible, remote work.
Work from anywhere in Japan with an internet connection.
Use specialized legal expertise in a rapidly growing technology field.
Help shape how AI handles complex legal questions.
Evaluate Japanese legal AI responses using your corporate and M&A expertise. This remote contractor role offers 20+ hours per week at $140–$150 per hour for qualified attorneys in Japan.
Use your active family law license and practical legal judgment to evaluate AI-generated legal content, create realistic training scenarios, and improve model accuracy. Remote contractor work offers 20+ hours per week at $140–$400 per hour.
Apply your immigration-law expertise to evaluate AI-generated legal content, refine documents, and improve how AI handles complex legal concepts. This worldwide, part-time contract role offers 20+ hours per week and pays $140–$400 per hour.