Yes. Health AI training and evaluation work is carried out remotely, in your own time, at times of your choosing.

Do I need technical or coding skills?

No. Clinical judgement and real-world experience are what is required. Most tasks involve reading, reasoning, and writing assessing AI outputs, defining reference answers, and quality checking responses.

Is patient data involved?

No identifiable patient data is used.

Is this a job or freelance contract work?

This is contract project work, not employment. Most arrangements are flexible and task-based rather than salaried. You will be contracting directly with Mercor or Micro1, not with this site.

Can I do this alongside NHS or locum work?

In many cases yes subject to your employer's policy on secondary work and any conflicts of interest.

Maximising Your Potential: Assessments, Certifications, and Free Training

By Sean Key, Editor — Applied Clinical Judgement. Last reviewed July 2026.

Once you are in a platform’s pool, the amount of work you can be matched to is not fixed. It grows with the skills you have verified. Every additional skill you certify widens the range of projects you are eligible for, and higher assessment scores tend to unlock better-paid work. This page is about using that deliberately – turning your profile from eligible-for-a-few into eligible-for-many.

It also covers the free training the platforms provide, which does two things at once: it teaches you the work, and it signals that you are serious.

More verified skills, more matches

The principle is simple and both platforms state it plainly. On Micro1, certifying additional skills increases the range of projects you may be considered for – each verified skill is another door. On Mercor, completing the platform’s assessments is part of what makes you matchable, and your interview and assessment performance feeds into which work you are surfaced for.

So the move is to certify or assess the skills you genuinely have but have not yet validated. If you can do model evaluation and translation and domain-specific review, get all three verified rather than leaving two of them implied. Do not fake expertise you lack – the assessments are designed to catch it, and a failed one wastes an attempt. But do claim, and prove, the real skills you have been leaving on the table.

The generalist-expert route, and the skills it needs

Not all of this work is specialist. A large and steady part of it is general evaluation: judging whether an AI’s answer to an everyday question is accurate, sensible, safe, and genuinely helpful – across whatever topic comes up. These generalist roles are worth understanding, because they are often the most accessible entry point and they draw on the wide, ordinary knowledge covered in the CV guide rather than a narrow specialism.

The skill these roles reward is not subject expertise so much as good judgement, applied carefully and consistently. In practice that means:

Reading closely and spotting what is wrong. Catching the error, the missing caveat, the confident-but-incorrect claim – the things a rushed reader would miss.
Following a rubric exactly. Evaluation work runs on scoring guidelines, and the value you add is applying them consistently rather than substituting your own taste. The best evaluators are calibrated, not contrarian.
Explaining your judgement clearly. It is not enough to mark something wrong; you have to say why, in writing another person could follow. Clear, specific reasoning is the actual product.
Staying consistent. Scoring the hundredth item to the same standard as the first is harder than it sounds, and it is exactly what these projects are paying for.

These are learnable skills, and they improve with the assessments and practice below. If you are strong on careful reading and clear written reasoning, generalist evaluation work is often the fastest way in – and the breadth on your CV is what surfaces you for it. One thing worth knowing: because these generalist roles are common and recur often, they are also the ones most likely to be filled from the existing pool before they are widely advertised, which is a strong reason to get yourself into the pool early rather than waiting.

This is also why the whole hiring process tests what it does. Following a brief exactly, staying on the point, being concise, and writing clearly are not hoops the platforms make you jump through for their own sake – they are this job, described above. The application, the interview, and any written sample are all assessing the same habits you will use every day in the work. Seen that way, preparing for the assessments and getting good at the work are the same task.

How scores affect your work and pay

Assessments are not just pass-or-fail gates; the score can matter. On the platforms that use scored vetting, a higher score tends to unlock higher-tier, better-paid projects, and scarce skills command premiums because fewer people can prove them. So it is worth preparing properly for an assessment rather than taking it cold – the difference between a pass and a strong pass can be the difference between routine work and premium work.

Consistent quality once you start also compounds. Reliable, high-quality output on early projects tends to lead to more work, better rates, and priority access to premium tasks. The first few projects are, in effect, another assessment.

Use the free training

There is genuinely useful free training available, from the platforms themselves, from ACJ, and from the wider ecosystem. Using the official material has a second benefit: it cannot get you flagged, unlike third-party tools. Here is what to use and where to find it.

The platforms’ own training and practice.

Micro1’s free course. Micro1 publishes a free course, “Inside micro1: How human expertise powers AI”, hosted on Udemy – around half an hour, no AI background needed. It covers what the work is, how models are trained, and how your feedback shapes them through RLHF. Search for it by name on Udemy, or reach it through Micro1’s own site. If you are new to AI training work, do this first: it means you understand the job you are being assessed for.
Micro1’s Interview Prep tool. A free tool that simulates the real Zara interview and gives feedback. You will find it in Micro1’s interview or help section; it is the best preparation for their interview there is.
Mercor’s Practice Interview. Mercor offers a free practice interview that does not affect your applications, reachable from your Mercor dashboard or their talent documentation. Rehearse on it before the real thing.
Mercor’s talent documentation. Mercor publishes guidance on the interview, assessments, and how matching works, in its talent docs. It is worth reading before you apply, because it tells you exactly what the platform checks.

The practice tools are covered in more detail on our interview guide, alongside the important warning not to use third-party AI copilots, which the platforms detect and penalise.

ACJ’s own free training resources. ACJ maintains a free page of curated, low-barrier AI courses, currently oriented towards clinicians and health AI but a useful starting point for anyone wanting to understand how AI is trained and evaluated before they apply. You can find it at ACJ’s free AI training page. The courses listed there are chosen to be genuinely accessible, with free access to the learning content, and completing some of this background can strengthen your application.

Certification pathways. Some platforms offer structured, free routes from fundamentals to certified evaluator. Where they exist, they are both preparation and a credential, so take them when they are offered.

A sensible order

If you are starting out: do the free course to understand the work, use the free practice tool to get comfortable with the interview, pass the interview, then certify every real skill you have to widen your matches. If you are already in a pool and quiet: certify additional skills and make sure your assessments are complete – that is often what turns a slow dashboard into a busy one.

One caution on sequence: do not wait until you have certified everything before you apply. Get in first with your core skills, then widen from inside the pool. The extra certifications add matches over time, but they are no reason to delay your first application – places can go to whoever qualifies first, so applying sooner with a solid profile beats applying later with a complete one.

Frequently asked questions

Will certifying more skills actually get me more work?

Yes – that is the point of it. Each verified skill widens the range of projects you are eligible to be matched to. Certify the real skills you have not yet validated.

Do assessment scores affect my pay?

Often, yes. On platforms with scored vetting, higher scores tend to unlock higher-tier and better-paid work, and scarce skills earn premiums. Preparing properly for an assessment is worth it.

What free training should I do first, and where do I find it?

If you are new, start with Micro1’s free half-hour course, “Inside micro1”, on Udemy, then use the platforms’ free practice interview tools – Micro1’s Interview Prep tool and Mercor’s Practice Interview, both reachable from their own sites – before the real thing. ACJ also keeps a free page of curated, accessible AI courses that is a good grounding, currently focused on health AI but useful more widely. All of these are official or curated, and none can get you flagged the way third-party AI tools can.

I am already in a pool but getting little work. What helps?

Certify additional real skills and complete any outstanding assessments – widening your verified skills is often what opens up more matches. Keeping your profile and availability current matters too.

Do I need to be a specialist, or is there generalist work?

There is a lot of generalist evaluation work – judging everyday AI answers for accuracy and sense – and it is often the easiest way in. It rewards careful reading, clear written reasoning, and consistent rubric-following rather than deep subject expertise. A broad CV is what surfaces you for it.

The bottom line

Your matchable work grows with your verified skills, so treat assessments and certifications as the lever they are. Do the free training to learn the work and signal you are serious, certify every real skill rather than leaving it implied, and prepare properly so your scores open the better-paid tiers. The more you can prove, the more you get offered.

This guide covers the last stage of getting hired. For the whole path – CV, LinkedIn, interview, and getting matched – see our complete getting-hired guide.

—

Applied Clinical Judgement is a referral service run by Sean Key, a Digital Health Senior Programme Manager with 29 years of NHS and private-sector experience. [LinkedIn] · [Book a vouching call] · [Application guide (PDF)]. ACJ is not an employer or recruiter; it points you to the platforms that do the hiring, and may receive a referral credit at no cost to you.