DataAnnotation vs micro1 vs Mercor vs Ethos vs Handshake AI vs Alignerr vs Outlier (2026)

By Luke Lashley · Published July 13, 2026 · Updated October 9, 2026

Some links here are referral links. If you sign up and do paid work, the platform may pay me; it costs you nothing. Full disclosure.

I’ve applied to all seven of these platforms. Five have paid me for work, nearly $30,000 between them since June 2026. Ethos hired me, and my first payout there lands October 14. Outlier has paid me $0.

Here’s how I rank them, going by what I’ve been paid and how much work they’ve given me:

• Worth your time: micro1 ($12,408.75) and DataAnnotation ($14,263.79). Both have paid me for a lot of real project work.
• Strong, with a catch: Ethos. It hired me at $100 an hour after a short voice interview, then my client moved the start date.
• Good pay, uneven hours: Mercor ($1,423.88). My two contracts pay well, but the work comes in short sprints with long gaps.
• Only for the right role: Handshake AI ($919.99) and Alignerr (about $860). Good rates, but projects can end fast and the pay rules need a close read.
• Skip it: Outlier. The screening was long and unpaid.

Every figure here comes from my own dashboards and payout emails.

The short version

Who each platform is for

For programmers, DataAnnotation is my top pick. There’s usually $50-an-hour coding work on my dashboard, though it’s slowed down a bit lately. I’ve worked at up to $75 an hour, and my best month was August, at $4,726.68. The tasks are often fun. The assessment took me about an hour, and I was accepted about a week later. All of that is from a coding account; some generalists report dry dashboards.

I’ve sent micro1 17 applications and landed three projects, at $80 to $90 an hour. While one is running I usually work 10 to 20 hours a week. They’ve lasted anywhere from a few days to about three weeks, and the third is still running in October. The hard part is the Zara interview. I failed my first one.

Ethos hired me as a software engineer at $100 an hour. Each role I applied for came with an AI voice interview of 15 to 20 minutes. Onboarding meant an ID check and an offer letter, then a paid practice task and the client’s paid screening task. My first payout, $306.67 for those two tasks, is due October 14. The client has moved the project start, and the latest word is it starts within days, so I can’t tell you yet how steady the work is.

My two Mercor contracts pay $95 and $85 an hour, and both were instant offers for roles I hadn’t applied to. The roughly 15 roles I did apply to never turned into a contract. The catch is volume: I’ve billed 7.88 hours in total. $700 of what Mercor has paid me was incentives, starting with a $400 onboarding bonus and then completion bonuses per task.

Getting into Handshake AI took a short form, with no AI interview. My June project paid $100 an hour, and I made almost $400 in one week before I was removed for “changing project needs.” Now I’m on a $125-an-hour project that’s waiting for the client to kick off. It’s paid me $540 in incentives so far, including a paid onboarding call and a trainer workshop, and I get $500 more if I pass the client’s screener, all before the project’s first task. Handshake has some of the highest- and lowest-paid roles I’ve seen, and a high chance of being offboarded. If you can go with the flow and take what you can while it lasts, it can be pretty good.

Alignerr is worth an application if you have a matching specialty, but I’d apply to micro1, DataAnnotation, Ethos and Mercor first. You get about seven application slots, so spend them on roles that fit. My project was advertised at $80 to $120 an hour and paid me $75 to $95. The pay rules changed a lot. At one point it was $75 an hour and $0 for a task that failed, which is how one four-hour task paid me nothing. Later it was $80 an hour for completed tasks and $90 if they were accepted. All my qualifying work has been paid. I found the project tedious and the project managers quiet, and I was removed from the current phase on October 7, pending the next one.

Outlier has paid me nothing. The screening was long and tedious, and it didn’t lead to work. I’d only reconsider it for a project that matched my skills exactly.

Head to head

DataAnnotation vs micro1

micro1 pays me more per hour ($80-$90 against $50-$75), but you apply role by role and projects end. DataAnnotation is where I go when I just want to work, because there’s usually something on my coding dashboard. In total they’re close, with DataAnnotation a little ahead. I keep both.

Mercor vs micro1

Mercor pays more per hour. micro1 has paid me far more in total, because that’s where the hours are. My micro1 projects ran 10 to 20 hours a week. On Mercor my best week was 2.43 hours.

Ethos vs Mercor vs micro1

For me, Ethos had the shortest screening of the three: one voice call per role, and the screening tasks pay. Mercor and micro1 both interview you on camera, and I failed my first Zara interview. Mercor’s AI interviewer is much worse than micro1’s, in my experience: pause too long and it thinks you’re done talking. Where Ethos loses, for now, is that I haven’t worked a project hour there yet. My Ethos review has the full timeline.

DataAnnotation vs Outlier

For a programmer this one isn’t close. DataAnnotation pays me $50 an hour or more for coding work. Outlier’s screening took real time and paid nothing.

Outlier alternatives

If Outlier put you off, the rest of this list has treated me far better. Five of the other six have paid me for work, and Ethos has my first payout scheduled. The only Outlier rate I saw was a listed $15 an hour, on one project card.

Start with micro1 and DataAnnotation, which have paid me the most. Add Ethos if you have several years in a professional field, and take Mercor’s assessments. Handshake AI is the quickest application to send, and Alignerr fits when a role matches your specialty. If you live outside the US, the country notes below cover who each platform takes.

Which countries each platform takes

Where you can be hired and where you can be paid are two different lists. Here’s what the platforms publish:

• micro1 says it can’t hire residents of 21 countries and regions: Afghanistan, Belarus, China, Cuba, DR Congo, Ethiopia, Hong Kong, Iran, Iraq, Libya, Macau, Myanmar, North Korea, Russia, Somalia, South Sudan, Sudan, Syria, Ukraine, Venezuela and Yemen. The Philippines, Pakistan and India aren’t on it. Some roles are narrower, such as US-only or US, UK and Canada.
• Mercor publishes a supported-countries list, 107 countries as of October 9, 2026, and can’t onboard people anywhere else.
• Ethos pays through Wise or Revolut, and the payout dropdown I see lists about 160 countries. That covers payouts only; it doesn’t mean Ethos hires in all of them. Its expert terms say the payout account must be in your own name.
• DataAnnotation pays by PayPal, and PayPal’s ability to receive money varies by country, so make sure it works where you live.

Individual listings can be narrower than any of this. A listing that says “remote” without naming countries isn’t the same as worldwide.

The order I’d apply in

I apply to the quick ones first and keep looking while applications are pending. Before you stack projects, look over each contract. My guide to applying covers resumes and timing, and the job finder shows what’s open right now.

1. Handshake AI. My application was a short form based on my resume and profile. Projects can still require assessments or a client screener, so before you accept one, look at how it pays and what its time cap and approval rules are. Apply to Handshake AI →
2. Ethos, if your field is listed. Each role’s screen is a short AI voice call. See open Ethos roles.
3. DataAnnotation, if you code or have a testable specialty. My coding exam took about an hour. Generalists can leave this one for later. Read my DataAnnotation review and apply.
4. Mercor’s talent networks and assessments. Join the talent networks and do the assessments. They’re per skill, so a few hours carry over to many roles, and people often get an instant offer for a different role than the one they applied to. Start on Mercor →
5. Most of your time on micro1. The first Zara interview is the one to prepare for; later applications can reuse qualifying results from the past year. My projects paid $80 to $90 an hour, for very different lengths of time. Apply to micro1 →
6. Alignerr last, with roles that fit. Fill your slots, then wait. An approval or a rejection frees a slot, and a stale application is worth swapping for a fresher one. Pick an Alignerr role →

Outlier only goes on the list if one of its projects matches your skills closely.

All of this is side income for me. My steadiest platform still has quiet weeks, so I keep several going at once.