HiringPart-time OKRemoteSteady work
RLHF Feedback Rater
Northlight AI · Remote (US/EU) · Contract · Remote
$22 – $34 / hr
Why this role holds up next to AI
Models learn what 'good' means from humans ranking their answers. Your taste literally trains the next generation of AI.
What the work is
Compare AI answers and pick the better one, over and over, to teach models human preferences. Steady, remote, part-time friendly.
Day to day, you'd
- Rank pairs of AI responses by quality and safety
- Apply detailed rubrics consistently across many tasks
- Surface edge cases the guidelines don't cover
You're more ready than you think
You already do this if you were a grader, reviewer, or detail-oriented analyst.
Apply for this role
Create your free account to apply — it takes under a minute.