Rate, label and correct AI model outputs on platforms like Outlier, Scale and Surge — no portfolio required.
Why that grade Based on an analysis of 1,500+ Reddit posts giving real effective hourly rates by task type — a large, self-reported sample, not audited platform payroll data. Course-seller index 3/10.
Figures are researched estimates, not guarantees. Check local rules before you trade.
AI labs need enormous volumes of human judgment to train and evaluate models, and this is the lowest-barrier entry point into that work — no portfolio or credential is required for the generalist tier, though specialized skill pays meaningfully more.
Bulk micro-annotation $10-16/hr; standard RLHF/conversation review (most common/available) $18-28/hr; long-form content evaluation $20-32/hr; code review/debugging (verified coding background) $28-45/hr; expert-tier (legal/medical credentials) $35-60/hr. One cited real example: $687 for 31.5 hours (~$21.81/hr).
Someone who wants flexible, no-commitment supplemental income and can tolerate unpredictable task availability.
Anyone needing predictable, stable primary income — project availability fluctuates sharply and isn't something you control.
Task availability swings from full queues to nothing for days with no warning, and there have been documented payment-processing delays and account-deactivation issues on some platforms in 2025-2026, so this cannot be relied on as stable income.
6 steps to your first paying customer
Get the basics in hand before you take on anyone: A computer and reliable internet.
Done when You have a working computer and internet connection confirmed against each target platform's minimum requirements.
DataAnnotation/Surge and Outlier/Scale at minimum — project availability on any single platform is unpredictable, so diversify from day one.
Done when You've submitted applications to at least DataAnnotation/Surge and Outlier/Scale on the same day.
The rest of the playbook: what to charge, what you need in place before you take money, where the first customers come from, and what each step costs.
Free forever · no card · 30 secondsN/A — this is gig work, not a business with a moat to defend.
Building a strong platform track record/score over time for access to better projects
Developing specialized skills (coding, STEM, languages) that pay a real premium
None — treat this as income, not a sellable asset.
Same idea, different rules. One playbook, with the facts that actually differ overlaid per market.
Decent access but fewer high-paying expert-tier projects than the US.
Access to generalist tasks is strong; expert-tier access is more limited.
Widest access to the best-paying specialized and expert-tier projects.
Get matched to 30-90 minute paid interviews via Respondent.io and User Interviews.
Take part in university-run behavioral and psychology studies for guaranteed minimum hourly pay.
Test websites and apps and narrate your reactions for UserTesting.com and Prolific, for pocket money.