What our experts do
Companies building AI for regulated work hire us to test it independently. Our experts are the standard those tests are measured against.
Review outputs
Read AI-generated advice, assessments, and documents, and judge them against a clear rubric, the way you'd judge a junior's work in your own field.
Write the reasoning
A short written judgement with each review: what's wrong, why it matters, and what a correct answer needed. Your reasoning is what sets the standard.
Shape the tests
Experienced panel members help define what we test: the realistic scenarios and edge cases where plausible answers go wrong.
Time & pay
Project-based, paid fast.
- Work comes in batches per project: typically a few hours, fully remote, on your schedule
- Paid per item or per project, agreed before you accept the batch
- Invoices paid within 7 days, not 30, not 60
- No minimum commitment; accept or decline each project
Standards
Independence is the job.
- Conflicts of interest declared before every engagement: if you work with or against the company being tested, you sit it out
- Everything you see is confidential; everything you write is anonymous in our reports unless agreed otherwise
- Selection is a short paid calibration exercise: we measure judgement, not CVs
- You must hold a current credential and be in active practice
Practising professionals in finance, law, and healthcare: the people whose standard AI in these fields is supposed to meet.
Solicitors and barristers. Accountants, compliance and risk professionals, paraplanners and advisers. Doctors, psychologists, therapists, pharmacists, and allied health practitioners.
If your judgement carries professional accountability, it's the judgement we need.
Apply to join