Remote Web Researcher

Turing • San Francisco, CA

23h • Remote

About The Position

Turing is hiring a Web Researcher to help build an evaluation benchmark for frontier AI browsing agents. The role focuses on designing research problems that state-of-the-art AI systems cannot solve reliably, even with full web access and multiple attempts. It is investigative research rather than subject-matter expertise or content writing: begin with a verifiable fact, work backward to create a difficult question, and substantiate the result with a complete, auditable evidence trail.

The researcher will produce natural-language questions with short, stable, objectively verifiable answers; independently checkable clues spanning multiple fact types and constraints; and validation records documenting obvious searches and their results. The evidence trail is the majority of the work. The assignment requires 40 hours per week, including at least four hours of overlap with PST, and lasts eight weeks. It is a contractor assignment with no medical or paid leave. Applicants complete an assessment, which the team evaluates; selected candidates are contacted about next steps.

Requirements

• Demonstrated open-web research ability, including locating primary records and navigating government and institutional databases, archives, registries, and PDF documents.
• Precision with sourcing, including citations to exact pages, tables, and sections rather than just homepages.
• Comfort researching unfamiliar subjects from scratch.
• Native or near-native written English.
• High tolerance for structured documentation.
• Experience with LLM evaluation, red-teaming, or benchmark construction.
• Experience in at least one of these domains: reference librarianship, archival research or special collections; investigative journalism or professional fact-checking; OSINT, due diligence, KYC or investigative research; patent, prior-art or legal-discovery search; genealogy and records research; competitive quizzing or puzzle-hunt construction.
• Availability for 40 hours per week with at least four hours of PST overlap.

Nice To Haves

• Familiarity with JSON and structured data delivery formats.

Responsibilities

• Design research problems that are difficult for frontier AI browsing agents to solve, starting from a verifiable fact and working backward to construct the question.
• Create natural-language research questions with short, stable, objectively verifiable answers.
• Develop independently checkable clues spanning multiple fact types, including dates, people, places, organisations, works, events, records, and quantities, with specific constraints.
• Provide a complete, auditable evidence trail, citing exact pages, tables, and sections.
• Create validation records showing obvious searches performed and the results returned.
• Complete the application assessment.

What This Job Offers

• Education Level: No Education Listed