If you’re considering clinical AI training work, this onboarding process can feel opaque from the outside. It isn’t. Several steps are automated, applications are reviewed in batches, and the actual work varies week to week.
The main thing to understand up front: you’re not applying for a single job. You’re joining a talent pool. Each stage below filters for a different kind of fit, and platforms disclose Scotland certificate and identity verification during the process, so expect that step.
Who this is for: regulated healthcare professionals applying to platforms such as Mercor or Micro1 for clinical AI training. AI output evaluation, prompt review, gold-answer creation, and safety or domain-expert oversight.
The onboarding stages
Stage 1: Application and eligibility screening (minutes to a few days)
You submit your CV or LinkedIn profile, confirm your clinical background and provide any credentials or registration information the role requests. Identity verification can occur during account setup, application or later onboarding — typically via a mobile app: passport scan, face photo, chip verification. A formal background check, where a platform or client requires one, may happen later — often after matching or during project onboarding — so do not assume it always occurs at the initial application stage. In my own case Mercor posted a Disclosure Scotland certificate a few days after I applied, but treat that as one experience rather than the standard timing. At this stage, platforms are checking eligibility and completeness, not ranking you against other candidates.
Stage 2: AI-led interview and assessment (30-60 minutes)
This is the most consequential stage and where most filtering happens. Expect scenario-based questions, evaluation of AI-generated responses, and boundary-setting. How these interviews are structured and scored is covered in the companion guide below.
Stage 3: Domain or task-specific evaluation (days to a few weeks)
Some roles require additional validation: reviewing example outputs, writing or refining rubrics, applying a scoring rubric, or explaining why a given answer is unsafe or incorrect. This is more common for senior or specialist reviewers.
Stage 4: Human review and verification (days to a month, not consistent)
Where it happens, this is usually a spot check of AI-scored results, credential verification, or a short clarifying message, not a traditional interview or a validation call.
Stage 5: Pool acceptance (immediate to a few days)
You complete the agreements and compliance steps required for the platform or project. Depending on the process, contract, payment and verification steps may occur before or after this point. Once the required stages are complete, you are eligible to start the project when work is released.
What happens after approval
Work is matched to your background, prior project fit, domain relevance, performance scores, and availability. Early on, expect gaps and calibration tasks as the platform builds confidence in quality before scaling volume.
Time to first paid task
| Fastest | 1-2 weeks |
| Common | 3-6 weeks |
| Sometimes | Several months |
Silence doesn’t mean rejection. Most clinicians who do well see an uneven start before things settle into a rhythm.
Possible outcomes
| Approved and working | Regular or semi-regular tasks. Performance feedback influences volume over time. |
| Approved but quiet | Passed onboarding, few tasks initially. This reflects market demand, not your ability yet. |
| Conditional approval | Approved for certain task types only. Scope can expand with good performance. |
| Feedback varies by outcome. On micro1, a Did Not Qualify result can currently include a Request Detailed Feedback option in the Talent Dashboard. A Not Selected outcome can mean you met the role criteria but another candidate was chosen, and personalised feedback may not be available. See Getting Matched for the current status definitions and check date. | Mercor currently provides no feedback. Micro1’s rejection email can be replied to for feedback, genuinely useful when it comes. Common reasons include safety or boundary issues and inconsistent answers. Micro1 currently allows re-application after 30 days. |
How to approach this psychologically
- Portfolio mindset: freelance expert work, not employment.
- Quality over speed: careful judgement beats fast answers.
- Expect gaps: variability early on is completely normal.
- Silence isn’t failure: demand-driven timelines cause quiet pauses.
Read next: How to pass the AI interview, or practical preparation for the technical setup.