Turing · via Braintrustnew this weekWeb Researcher - AI Training
Good fit for: data annotators, labelers, and detail oriented generalists.
Listing verified on the platform’s official board. Last checked September 25, 2026.
About the project
Create well-sourced research tasks that help evaluate and improve AI systems. This work combines advanced web research, precise writing, source verification, and thoughtful task design.
What you will do
- Develop concise research prompts supported by clear clues and reputable, publicly accessible English-language sources.
- Verify answers and record citations, source links, supporting evidence, and research paths.
- Check source accessibility and accuracy, evaluate AI-generated claims, and revise tasks following quality feedback.
- Protect confidential project information and follow project standards.
What you bring
- A master's degree and at least three years of relevant experience.
- Demonstrated sourced research experience, such as journalism, fact-checking, library or archival work, due diligence, graduate primary research, or citation-focused editing.
- Fluent English, strong writing, advanced web research, and careful source evaluation.
- Domain expertise, OSINT, cataloguing, or experience creating benchmarks, exams, or puzzles is helpful, not required.
Project details
Remote freelance project, expected to last six weeks. Availability of 20–40 hours per week depending on project availability, including four hours of overlap with Pacific working hours. Compensation is USD 30 per approved task, with an average duration of 3 hours per task.
Applying via Braintrust: Braintrust is a zero-fee talent network — you keep 100% of your rate. This role is posted by the listed company; apply through Braintrust.