Frontier AI Research Investigator | Web OSINT Benchmarking

Turing; Seattle (WA); On-site; USD 83,000 - 165,000; Full time

Benefits offered by this job

JSON familiarity

Job summary

Turing is building an evaluation benchmark for frontier AI browsing agents. You will design research problems that start from a verifiable fact and require locating hard-to-find information, then prove your work with an auditable evidence trail.

The role emphasizes investigative research, not subject matter expertise or content writing, and requires precise sourcing and open-web research in English.

Qualifications

• Open-web research ability with primary sources.
• Cite exact pages, tables, and sections - not just homepages.
• Native or near-native English writing.
• Experience with LLM evaluation, red-teaming, or benchmark construction.
• Comfort researching unfamiliar subjects from scratch.
• Ability to locate records in government or institutional databases.

Responsibilities

• Design research problems that start from a verifiable fact and are hard to locate.
• Create a natural-language research question with a stable, verifiable answer.
• Develop clues spanning dates, people, places, organizations, works, events, records, quantities.
• Provide a validation record showing searches run and results returned.
• Maintain an auditable evidence trail documenting methodology.

Skills

Open-web research

Sourcing precision

LLM evaluation

Red-teaming

Benchmark construction

OSINT

Written English

Tools

JSON familiarity

Job description

Turing is building an evaluation benchmark for frontier AI browsing agents. You will design research problems that start from a verifiable fact and require locating hard-to-find information, then prove your work with an auditable evidence trail.

The role emphasizes investigative research, not subject matter expertise or content writing, and requires precise sourcing and open-web research in English.