micro1 and Mercor are the two largest sources of AI training roles on this site. Together they account for 494 of the 597 open roles we track. But they are different kinds of platforms, and the work — and pay — reflect that.
All figures come from each platform’s own listings, checked this morning.
The numbers
| micro1 | Mercor | |
|---|---|---|
| Open roles | 234 | 260 |
| Stated openings | 59,574 | 8,815 |
| Hourly pay range | $6–$400 | ~$13–$270 |
| Pay unit | Hourly | Hourly |
micro1 has fewer roles but far more openings per role. A single micro1 listing can carry 5,000 or 10,000 openings because it represents a large-scale annotation or evaluation programme hiring many people for the same task. Mercor roles tend to be smaller — often tens or low hundreds of openings — because the work is more specialised.
How the work differs
micro1 covers the full spectrum. Its roles range from generalist data labelling (no domain expertise needed, high volume, lower pay) to specialist expert review (deep domain knowledge required, lower volume, higher pay). The generalist tier is where micro1’s scale comes from: thousands of people doing the same annotation task.
Mercor skews toward professional and technical work. Its listings tend to be for software engineers, data scientists, analysts and subject-matter experts. The tasks are more likely to involve writing code, reviewing technical outputs, or applying professional judgement to complex problems. Mercor’s floor is higher because it posts fewer of the high-volume, low-rate roles that bring micro1’s floor down.
Pay
The ranges overlap, but the centres of gravity are different.
micro1’s bulk sits in the $15–$75/hr range for most roles, with outliers at both ends. The $6 floor is a high-volume annotation task; the $400 ceiling is a niche expert review. Mercor’s rates cluster in a similar band but start higher — there is no $6 data labelling tier.
Neither platform guarantees hours. A listing says what the rate is, not how many hours of work are available. A $75/hr role with two hours of available work a week pays less than a $25/hr role with forty.
Screening
Both platforms screen applicants, but the process differs.
micro1 runs a skills assessment tied to the specific role: a coding challenge, a writing sample, a domain quiz. The assessment can take thirty minutes to a few hours. Once you pass, you are matched to work as it becomes available.
Mercor often includes a technical assessment and may add an interview or work sample. For its higher-paying roles, screening can be more involved.
Neither platform charges applicants.
Which one should you try?
If you are a generalist — good at following guidelines, comfortable with repetitive evaluation work, looking for volume — micro1 has more of that work and more openings to fill.
If you are a specialist — a software engineer, a domain expert, someone whose professional credentials are the reason a platform would hire you — both have roles worth looking at, but Mercor’s listings are more likely to match that profile.
If you are unsure, apply to both. The screening is free, the application is separate, and nothing stops you from working on both platforms. Browse the micro1 roles and Mercor roles to see what is open now.
For a detailed look at each platform’s terms, screening process, and what we do and do not know about them, see the micro1 platform page and Mercor platform page.