AI Benchmark & Red-Teaming Specialist

In short

Turing is hiring a full-time remote AI Benchmark & Red-Teaming Specialist to test and improve frontier AI models. You will help expose AI weaknesses through research and benchmarking, making this a great fit if you have a background in investigative research and want a flexible remote schedule.

What you will do

• Research niche topics to find hard-to-surface facts using advanced web search.
• Verify facts against primary sources and document exact URLs.
• Draft benchmark tasks designed to trigger model failures.
• Test tasks against large language models and analyze why they failed.
• Review benchmark tasks created by peer trainers.

What you need to apply

• 1+ years of experience in investigative research, fact-checking, AI benchmark authoring, red-teaming, or technical QA review.
• Intermediate-to-advanced web search and OSINT skills.
• Strong analytical capability to inspect model reasoning chains and debug failures.
• Ability to write clear benchmark prompts with primary-sourced answer keys.

Who can apply

This full-time contractor role is open worldwide with no country restrictions stated. You must be able to commit up to 40 hours per week with 4 hours of overlap with PST.