Training Turk

Interview experiences

What the real interview was like

Short accounts from people who sat AI-training interviews and tests on Mercor, micro1, Outlier and more.

Each one is told in our own words, from a post the candidate shared publicly.

Student contributor, software work for AI labs

Mercor · Dec 2025 · Software engineering

A recruiter offered a student paid remote work with teams that help AI labs train models. The 20-minute AI interview skipped small talk and went straight to backend depth and a system design case.

  • Community report, paraphrased
  • 1 round
  • 20 min
Read the full experience

How it ran

  1. A recruiter got in touch about a remote student role supporting AI lab projects.
  2. One 20-minute AI interview, purely technical, with a visible clock.
  3. Waiting for a decision.

What was asked

  • Backend architecture and how it was used in real projects.
  • How parts of a system talk to each other, such as choosing REST or GraphQL.
  • What was actually built during internships.
  • Designing a financial chatbot: latency, caching, WebSockets versus polling, several data APIs.
  • Security: encryption, tokens, rate limits, compliance logging, SQL injection.

What the interviewer or test was like

  • No warm-up: technical questions from the first minute.
  • Cared more about why a choice was made than about definitions.
  • Intense but fair. Admitting a gap and reasoning it through seemed to land well.

Tips

  • Use real project examples rather than textbook answers.
  • If you have not done something, say so and explain how you would approach it.
  • Have concrete numbers ready, such as the traffic a system handled.

Outcome

Still waiting to hear back when the post was written.

Programmer, coding track

DataAnnotation · Jun 2026 · Software engineering

An experienced programmer took the DataAnnotation coding test, which took about three hours. The problems felt easy and were answered in detail, but no reply came two weeks later.

  • Community report, paraphrased
  • 1 round
  • 180 min
Read the full experience

How it ran

  1. Applied after seeing an advert.
  2. A coding test of roughly three hours.
  3. Two weeks with no decision.

What was asked

  • Review functions and explain which are right and which are wrong.
  • A puzzle-style programming problem.
  • Python was the main expected language.

What the interviewer or test was like

  • A self-paced written test with no live interviewer.

Tips

  • Set aside about three hours for the coding test.
  • Do not count on quick feedback.

Outcome

No response two weeks after submitting.

Coding contributor, many languages

Outlier · Jul 2025 · Software engineering

A contributor passed nine Outlier coding screenings in about a month, from Python to Verilog. No paid work followed at first.

  • Community report, paraphrased
Read the full experience

How it ran

  1. Signed up.
  2. Identity and phone verification.
  3. Nine separate coding screenings, one per language, over about a month.

What was asked

  • Screenings in Python, Java, Go, Dart, C++, C#, JavaScript, Swift and Verilog.

What the interviewer or test was like

  • Unpaid, self-paced screenings with no live interviewer.
  • Took a lot of time across many languages.

Tips

  • Passing a screening does not guarantee paid work.

Outcome

Passed all nine, but had no paid tasks when the review was posted.

Interviewed recently? Tell us how it went.

We summarise it in a few lines and never publish your name.

Practise for a roleShare your experience

Community reports are paraphrased from public posts and may be out of date. Platforms change their process often.