Remote QA Auditor for AI Evaluation Benchmarks

FutureSearch; San Francisco (CA); Hybrid; CAD 147,000 - 245,000; Part time

Benefits offered by this job

Fully remote

Job summary

FutureSearch is seeking a meticulous investigator to quality-check evaluation tasks for training and benchmarking frontier AI models. You will verify claims, cite sources, and ensure a verifiable rubric accompanies every task.

The role is fully remote, with flexible hours, a 1-month contract, and an expectation of ~20 hours per week. Prior hands-on AI tool experience is not required as we will train you.

Qualifications

• Must be obsessively detail-oriented and able to verify claims in academic papers.
• Able to cross-reference financial documents and sources to confirm correctness.
• Strong written communication; concise, accurate reporting of findings.
• Judges context and decides when to pull in teammates for review.

Responsibilities

• Quality-check evaluation tasks and verify correct answers with evidence and rubrics.
• Explain precisely what is wrong and why when findings fail.
• Review submissions under tight delivery deadlines and prioritize issues.
• Collaborate remotely with a global team across locations.

Skills

Attention to detail

Skeptical mindset

Written communication

Cross-referencing

Deadline-driven

Job description

FutureSearch is seeking a meticulous investigator to quality-check evaluation tasks for training and benchmarking frontier AI models. You will verify claims, cite sources, and ensure a verifiable rubric accompanies every task.

The role is fully remote, with flexible hours, a 1-month contract, and an expectation of ~20 hours per week. Prior hands-on AI tool experience is not required as we will train you.