Indonesian AI Evaluation Specialist - Remote & Flexible

aitrainer; United States; Remote; USD 12,000 - 21,000; Part time

Job summary

Handshake is seeking Bahasa Indonesian speakers based in Indonesia to help personalize AI responses. You will share real-world experiences using ChatGPT and evaluate AI responses to help future models tailor answers for individual users.

This is a project-based, remote, part-time opportunity with flexible hours and no minimum hour commitment. You will provide clear written feedback and compare responses to articulate which answers are better suited to users.

Qualifications

• Location — Based in Indonesia and able to work remotely in an asynchronous, self-directed capacity.
• Language — Fluent in Bahasa Indonesian, both written and spoken, with strong written communication; English (ideally B2) since materials are in English.
• AI experience — Prior hands-on experience using ChatGPT or similar AI chat assistants, with ability to compare responses.
• Judgment — Comfortable forming and defending personal preferences between two AI-generated responses.

Responsibilities

• Complete tasks on Handshake's annotation platform using your own real-world experience with AI tools
• Compare and evaluate AI-generated responses, sharing your honest preferences and reasoning
• Provide clear written feedback that helps AI better understand and personalize responses for individual users
• Complete tasks at your own pace with no set schedule or minimum hour requirement

Skills

Indonesia-based

Bahasa Indonesia

English proficiency

AI experience

Judgment / preferences

Tools

ChatGPT

Job description

Handshake is seeking Bahasa Indonesian speakers based in Indonesia to help personalize AI responses. You will share real-world experiences using ChatGPT and evaluate AI responses to help future models tailor answers for individual users.

This is a project-based, remote, part-time opportunity with flexible hours and no minimum hour commitment. You will provide clear written feedback and compare responses to articulate which answers are better suited to users.