Question 1

Is this really EU AI Act Annex IV compliant?

Accepted Answer

The dataset gives you the audit-ready ingredients: documented generation methodology, demographic balance, fairness baselines. Your conformity assessment is your responsibility — but you'd struggle to find a dataset that maps more cleanly to the Annex IV template.

Question 2

Why are names like Synthetic_Candidate_XXXXXX?

Accepted Answer

Explicitly synthetic naming eliminates any risk of identity collision with a real candidate. Document this in your audit log. If you need realistic-looking-but-synthetic names for UI demos, email us — we'll deliver a separate variant.

Question 3

Does salary correlate with gender or ethnicity?

Accepted Answer

Deliberately not. Salary is a function of years_experience, education_level, country, current_industry, current_company_size. Gender and ethnicity have zero predictive power on salary (verified — AUC ≈ 0.50). That's the fairness baseline.

Question 4

Can I use this to train a commercial ATS / matching engine?

Accepted Answer

Yes. The commercial licence includes perpetual derivative-works rights. Train models, embed in products, sell to customers. The only restriction is no redistribution of the raw dataset.

Resume data for AI hiring tools.
No real candidates.

Synthetic CVs — coming Q4 2026.

Specs

For whom

Resume data for AI hiring tools.No real candidates.

Synthetic CVs — coming Q4 2026.

Specs

For whom

Resume data for AI hiring tools.
No real candidates.