Web Research Task Author at Turing

Web Research Task Author | ~$30/Approved Task | Fully Remote | 6-Week Contract

Turing, one of the world's fastest-growing AI companies, is seeking skilled Web Research Task Authors to create rigorous, verifiable research tasks that evaluate how well AI systems navigate the open web. This is a fully remote contract role ideal for journalists, fact-checkers, librarians, archivists, or anyone with a strong background in primary-source research.

About Turing

Turing accelerates the advancement of powerful AI systems by partnering with leading AI labs to improve frontier model capabilities—spanning reasoning, coding, STEM, and more—while also building real-world AI solutions for enterprise clients.

What You'll Do

• Craft original 60–120 word research prompts that feel like genuine questions—no copied content, no embedded search hints.
• Identify four to six verifiable clues sourced from distinct, reputable, open English-language websites (excluding Wikipedia and paywalled content), each supported by a direct quote, URL, and verification date.
• Pinpoint a definitive, documented answer and clearly explain why alternative answers don't qualify.
• Document plausible wrong answers, identify which clue eliminates each, and articulate your full search path.
• Verify all sources while logged out of accounts; refine tasks flagged as too easy or unclear by reviewers.

Required Skills

• Advanced web research: search operators, reverse lookups, institutional catalogues, and specialized databases.
• Archive & catalogue familiarity: libraries, museums, legislative records, court records, sports/film databases, Wayback Machine, and preprint servers.
• Source evaluation: distinguishing primary from secondary sources, open from restricted, and stated facts from implications.
• Precise citation: extracting verbatim text from web pages, PDFs, scans, and tables with accuracy.
• Clue architecture: designing constraints that are individually searchable yet collectively pinpoint a single answer.
• Clear, fluent written English and comfort using both search engines and LLM tools—with a commitment to fact-checking every AI-generated claim.

Must-Have Qualifications

• Demonstrated sourced research background: journalism, fact-checking, library/archival work, genealogy, due diligence, graduate-level primary-source research, or citation-heavy editing.
• Availability of 20+ hours per week; any time zone accepted.
• Willingness to sign an NDA (materials are confidential client IP).

Nice to Have

• Deep subject expertise in history, science/technology, sports, music, politics, geography, film, or television.
• OSINT or cataloguing experience.
• Prior experience creating benchmarks, exams, or puzzle content.

Project Details

• Compensation: Approximately $30 per approved task
• Duration: 6-week project
• Hours: Up to 40 hours/week
• Location: Fully remote, worldwide
• Payment: USD
• Start Date: Upon successful completion of the assessment

To apply, submit your application and complete the required assessment. Candidates who pass will be invited to onboard immediately.