AIPayList

Live roles · by kind of work

AI model evaluation & RLHF jobs

Rating model responses, comparing outputs and flagging errors — the classic entry point into AI work. Most evaluation roles ask for strong reading judgment rather than credentials, which is why they sit below the expert tier; domain expertise moves them up.

255
live roles
$11–$37/hr
typical verified pay
$23/hr
median rate
$150/hr
top verified rate

● feed updated 56m ago · rates read from platform job feeds — “up to” ceilings never sorted as real pay (methodology).

What this slice looks like right now

  • ▪ Most of this work comes from Meridial (Invisible) (185), Mercor (29) and Mindrift (23).
  • ▪ Half of the advertised rates here sit above $23/hr, measured from 25 distinct advertised rates.
  • ▪ Best-paid role open now: SWE Infrastructure Specialist (Java) – Freelance AI Traine… $150/hr.
  • ▲ 23 listings advertise an “up to” ceiling, the highest $90/hr against a top verifiable $150/hr. We never rank on ceilings.
  • ▪ 23 roles were posted in the last 7 days.
  • ▪ 6 listings publish no pay at all. We show them rather than guess a rate.

See all 255 roles + filters →

Keep browsing