How to get started, in the order that wastes least of your time

Do the cheap checks first

Every platform here runs roughly the same five steps: you build a profile, you clear an eligibility gate, you sit an assessment, you complete onboarding, and then you wait to be matched. The steps cost wildly different amounts of your time, and most people do them in the worst possible order: sign up, sit an unpaid AI interview, and only then discover the project was never open in their country.

Reverse it. Each step below is cheaper than the one after it, so do them in this order and stop as soon as one fails.

Step 1: pick by profession, not by platform

These platforms are not interchangeable, and the fastest filter is what you already do for a living.

• A profession with a credential (medicine, law, psychology, engineering, consulting): micro1 and Mercor carry most of this work. Of the listings on this board, micro1 fields roles from BigLaw contract redlining to adolescent risk assessment, and Mercor runs project-shaped engagements with defined dates.
• A senior technical background: Alignerr runs engineering roles that build the evaluation infrastructure itself as well as judging its output.
• A degree but no niche specialism: Outlier recruits on declared subject expertise across a broad range, and requires at least an associate degree with specific projects asking for more.
• A decade or more in one field: Ethos sits at the senior end, matching credentialed professionals to advisory and AI training work rather than letting you apply to listings.

If your background is none of these, be realistic: this board is 22 senior roles out of 32. See what AI training work actually is for why the entry-level end of this market has thinned out.

Step 2: check eligibility before you spend an evening

This is the step people skip, and it is free.

Only two of the five publish anything you can check in advance. micro1 publishes a list of twenty countries and regions it cannot hire from. Mercor publishes the inverse: a supported countries and regions list, and states it does not engage contractors outside it. Alignerr, Outlier and Ethos publish no country policy we could find.

Then check the listing itself. Mercor puts a Location requirements field on individual listings and is unambiguous when it uses it: two of its current projects are scoped to the United Kingdom and say so in the title. micro1 is inconsistent: some listings say "Global, fully remote", some say "based in an English-speaking country" without defining which countries those are, and most say nothing.

The full breakdown, including the paperwork gates nobody files under eligibility, is in who can actually do this work.

Two things worth confirming while you are here, because they are quiet geographic filters: Outlier requires a valid ID and a mobile phone from your country of residence, and Mercor states it only engages people with the right to work and provide freelance services where they live.

Step 3: build the profile, because matching runs on it

On Mercor and Ethos you do not browse and pick. You apply once and are put forward for whatever fits, which means your profile is the entire product. On Outlier, the projects you see depend on the expertise you declare.

Write your profile as input to a matching system. Be specific about subfields rather than broad: "prostate cancer treatment sequencing" gets matched, "oncology" competes with everyone. Declare the credentials that gate the high-rate work, because several listings turn entirely on one: a bar admission, a completed residency, a PhD, a named firm on your CV.

Step 4: the assessment

Nearly every micro1 role routes through Zara, its own AI interviewer. It is unpaid and it happens before you know your rate; micro1's practice versions are labeled 18 minutes, and real interviews vary with the skills and exercises a role includes. Alignerr's own process page states that some jobs additionally require a Zara interview or domain assessments, so preparing once may serve you twice. Mercor runs its own AI interview, which it says usually takes about 20 minutes.

Two things about Zara are published and almost nobody uses them: there is a free practice mode generated for the specific role you are applying to, and there is a feedback report you can request if you are rejected. In a three-day sample micro1 published, there were 4,820 unsuccessful interviews and only about 10.7 per cent of those candidates asked for the report. Both are covered in the Zara AI interview, along with the five things worth doing before you sit it.

Outlier works differently: it puts onboarding at 30 to 90 minutes by its own estimate, including skill screenings, and then adds project-specific onboarding where you review guidelines and complete assessment tasks before any paid work begins.

Step 5: onboarding, identity and the payment rail

Only after the assessment does the administrative part start, and it is where a surprising number of applications stall.

Expect identity verification on every platform. Expect to set up a payment rail before you can be paid: Mercor uses Stripe or Wise, Alignerr uses Stripe Connect or PayPal, Outlier uses PayPal, Airtm or ACH bank transfer, and micro1 pays into a micro1 wallet. Expect tax paperwork: Mercor requires EU-based contractors to complete a W-8BEN during onboarding.

Confirm the rail reaches your country before you invest in the rest. Being eligible to be hired and being able to receive money are different problems, and only one of them is discussed on the listing.

Step 6: the wait

The platforms are more honest about this wait than the guides written about them.

micro1 states plainly that "certification does not guarantee immediate placement", and its own advice for the gap is to keep your profile current and certify more skills. Handshake AI states there is no fixed timeline for a first project invitation. Mercor matches you rather than letting you apply, so the wait is structural.

Treat the gap between clearing the assessment and receiving a first paid task as normal rather than as a rejection. It is worth applying across more than one platform for exactly this reason.

A realistic timeline

Nobody publishes end-to-end timings, so this is assembled from the steps that are published, and should be read as a shape rather than a promise:

The only stated turnaround we have found is micro1's note that roles typically fill within 48 hours, with a selected expert starting tasks 24 to 48 hours after onboarding. That describes how fast a filled role moves, not how long you wait to be picked.

What we could not establish

• No platform publishes an acceptance rate, so we cannot tell you the odds at any step.
• No platform publishes the typical wait between certification and a first task.
• Alignerr's practical country footprint comes from third-party reviews only and is attributed as such in who can actually do this work.
• Whether "based in an English-speaking country" is enforced by residence, by IP address, or by self-declaration is not stated anywhere we can find.

Sources

• The expert-facing pages and process documentation of micro1, Mercor, Alignerr, Outlier and Ethos
• Zara: An LLM-based Candidate Interview Feedback System, arXiv:2507.02869, micro1's own paper, for the interview figures
• micro1's interview prep page and Mercor's AI interview FAQ, for interview lengths
• The listings live on this site on 8 September 2026

Last updated 8 September 2026; interview lengths revised 25 September 2026. Platform processes change without notice, and a platform's own page always outranks this one. We are not affiliated with any of these platforms and we do not process applications.

Labeling Jobs does not give financial or tax advice.