FUTURESEARCH is looking for a meticulous investigator to quality-check the evaluation tasks we develop for training and benchmarking frontier AI models.
We are an elite team of engineers and researchers at the frontier of AI epistemics. We have the best publicly available forecaster, with a public track record in forecasting tournaments at evals.futuresearch.ai and markets.futuresearch.ai. We publish papers, such as our ICLR Workshop paper in Jan 2026, and benchmarks like Deep Research Bench and Bench to the Future. We work with frontier labs on research, evaluation, and training.
Our research pipelines produce evaluation tasks in overlapping waves, one batch in revision while the next is already underway. Each task is a hard research question of the kind in our public benchmarks, where a right answer exists but takes real digging to establish. Each ships with a verified correct answer, the evidence that proves it, and a scoring rubric, and every one of them has to survive independent review. You are the last check before that review. When something is wrong, you will say precisely what and why.
The job rewards a particular temperament. Fresh eyes catch what the person closest to the work overlooked, and the team will rely on you for exactly that. You ask the question everyone else skipped because the answer looked reasonable, and you review again to double-check. We work against real delivery deadlines, so part of the skill is judging what deserves a deeper look.
You do not need a research title, a technical degree, or prior hands-on experience with AI tools. We will get you up to speed on our tooling. Fact-checkers, editors, auditors, analysts, forecasters, and career QA professionals all fit this role, and so do people who have run high-stakes operations (payroll, compliance, cross-border paperwork) where one wrong detail becomes everyone's problem. If your title says operations or program management but your colleagues know you as the person who reads all the fine print, this role is for you.
Requirements
You are obsessively detail-oriented. If you’ve been accused of making perfect the enemy of the good, this role is for you.
You can patiently get to the bottom of a subtle verification problem, such as confirming a claim in an academic paper, cross-referencing financial documents, or tracking down an original source.
You are a calibrated skeptic. You can tell the difference between evidence that suggests an answer and evidence that settles it, and you notice when a source does not actually support the claim that cites it.
You communicate findings quickly, concisely, and accurately in writing. Just as important, you are able to judge how much context a teammate needs, which findings are worth pulling someone in on, and which you can resolve yourself.
We are fully remote. SF and London are our primary in-person locations. We have flexible hours.
The hourly rate for this role varies from $75 to $125/hour based primarily on location. You should expect to work an average of 20 hours/week for the duration of the contract. Week-to-week work will be dynamic, but expectations for hours will be communicated beforehand.
This is a 1 month contract with the opportunity to extend and/or convert to a full-time position.
Benefits
We are fully remote. SF and London are our primary in-person locations. We have flexible hours. The hourly rate for this role varies from $75 to $125/hour based primarily on location. You should expect to work an average of 20 hours/week for the duration of the contract. Week-to-week work will be dynamic, but expectations for hours will be communicated before-hand.
This is a 1 month contract with the opportunity to extend and/or convert to a full-time position.
.png)
