AI side hustles aren’t dead—they just got pickier. The bots still need babysitters, and they’ll pay you to judge them. 💸
My AI research agent pulled the receipts from papers and Reddit. The job is called RLHF: you rank model replies, write better ones, and flag safety issues. Those choices train a reward model; then the model gets nudged toward what humans prefer.
How the money works: expect an application, a tough exam, NDAs, and QA checks. Pay ranges from decent to “why am I doing this,” and task flow swings from feast to famine. Pro tip: don’t marry one vendor—keep 2–3 warm.
Quick hits from the trenches:
- TELUS: steadier search/eval hours if you pass a brutal exam; still cyclical.
- Surge AI: higher rates and real RLHF work; sporadic and selective.
- TaskUs: more like a job—stable schedules, metrics-heavy, location-dependent.
- Appen: after Google dropped a big contract, work got thin and choppy.
- Labelbox Marketplace: good for experienced teams; project-based, long waits.
- Sama: opportunities in emerging markets; past trauma-labeling controversies; pay often low.
- iMerit: technical projects; many roles tied to region/office; modest rates.
Here’s the catch: the “easy clicks” era is over. Tasks now are niche—safety QA, multilingual judging, coding/reasoning—with heavier compliance. Fewer generic gigs, more slow, careful work.
My take: if you want bread-and-butter hours, start with TELUS. If you’ve got strong writing/coding chops and can handle dry spells, Surge AI pays better. TaskUs is solid if you want stability over flexibility.
Tried RLHF as a side hustle yet? Which platform treated you best—and which one burned you?