Skip to content
Labeling Jobs

Domain Contributor - Evaluation Specialist

Pay
$30 – $60 / Hour
Open to
Worldwide
Apply

We earn a commission if you sign up through the links on this page. It costs you nothing and does not affect which jobs we list. How this works.

Skills
  • ai evaluation
  • source verification
  • research
  • task design
  • rlhf
  • written communication

What you'll do

You write tasks that push an AI model to its limits and then judge the answers. That means designing domain-specific questions and scenarios that need deep subject knowledge, assessing model outputs against authoritative primary sources, writing clear rationales for each shortfall (missed reasoning, source errors, missing nuance) and applying detailed guidelines and rubrics consistently. Deliverables go in independently on a rolling basis.

The ad frames the fields as history, law, finance, journalism and technology. It is flexible, pay-per-task work of about 15 hours a week, remote.

Who fits

The ad lists these as preferred qualifications:

  • Hands-on experience with AI model training, annotation, RLHF or evaluation (the ad says the role is for experts with prior AI training experience)
  • Deep expertise in a core domain: history, archival or library science, law (paralegal, legal academic, compliance), finance (equity research, PE/VC, forensic accounting), journalism, or technology and industry history
  • Advanced research skills and a record of verifying complex claims against authoritative sources
  • Strong written and verbal communication, and precision with detailed evaluation frameworks
  • Adversarial task design ("stomp the model") is advantageous
  • Graduate-level research, Fortune 100 experience or professional fluency in a non-English language are pluses

What it pays

micro1's listing states $30–60/hour. Compensation is output-based: you are paid per task that meets the project specification, the time per task varies with experience and workflow, and minimum submission requirements apply, including a minimum number of tasks per week that the ad does not specify.

Worth knowing

Good:

  • Open to several backgrounds, including paralegals, librarians and journalists
  • Adversarial task design is the interesting part of the job
  • The ad says roles are typically filled within 48 hours

Less good:

  • Prior AI training experience is expected, so this is not an entry route
  • Per-task pay and an unspecified weekly task minimum can make the effective hourly rate uncertain
  • You are expected to start within 24–48 hours of onboarding
  • The ad calls it an internal project and gives no project length

About this listing

Read on micro1's jobs board on 22 October 2026, shown as posted a day earlier. The band, hours and qualifications are the ad's published text. See micro1.

More roles at micro1

See all 323

Similar roles at other platforms

Guides about micro1

See all 58

Browse similar roles

Not the right fit?

See every open role, or get new ones on Telegram or Discord as they are added.