Investigative Web Researcher for AI Benchmarking

turing; United States; Remote; USD 83,000 - 124,000; Full time

Job summary

Turing is hiring for a contractor to build an evaluation benchmark for frontier AI browsing agents. Your work will involve designing challenging research problems that require an auditable evidence trail, starting from verifiable facts and reconstructing questions that make those facts hard to locate.

You'll deliver a structured output including a runnable set of clues across dates, people, places, organizations, works, events, records, and quantities, plus a validation trail showing search

Qualifications

• Proven open-web research ability locating primary records from government and institutional databases.
• Precision with sourcing—cite exact pages, tables, and sections.
• Native or near-native written English.
• Experience with LLM evaluation, red-teaming, or benchmark construction.
• Experience in archival research, investigative journalism, or OSINT is a plus.

Responsibilities

• Produce a natural-language research question with a stable, verifiable answer.
• Create independent clues spanning dates, people, places, organisations, works, events, records, quantities.
• Provide a validation record listing searches and results.

Skills

Open-web research ability

Primary source sourcing

English writing

Evidence trail documentation

LLM evaluation experience

Tools

JSON

Job description

Turing is hiring for a contractor to build an evaluation benchmark for frontier AI browsing agents. Your work will involve designing challenging research problems that require an auditable evidence trail, starting from verifiable facts and reconstructing questions that make those facts hard to locate.

You'll deliver a structured output including a runnable set of clues across dates, people, places, organizations, works, events, records, and quantities, plus a validation trail showing search