How EmpireO AI Matching Actually Works
A transparent look at how our AI scores candidates against vacancies — what signals it uses, how match percentages are calculated, and what improves your score.
Why we built it
International recruitment has a fundamental signal problem. A recruiter reviewing 200 CVs in a day can only extract a fraction of the relevant information from each profile. Skills are described differently across regions, qualifications have different names in different countries, and experience in one healthcare system doesn't obviously map to another.
Our AI matching was built to solve this — not to replace recruiter judgment, but to surface the right candidates faster and more fairly, regardless of how they describe their experience.
What the AI actually reads
When you apply or create a profile, the system processes:
- Job titles (current and historical) — normalised across naming conventions globally
- Skills and certifications — including synonyms (e.g. 'Staff Nurse' = 'RN' = 'Registered Nurse')
- Years of experience per specialisation — extracted from date ranges in work history
- Education level and field — with recognition of international institutions
- Location preference and visa status — matched against what the employer can support
- Sector-specific keywords — over 200 dimensions per role category
How the match percentage is calculated
The match score is a weighted composite. Weights vary by role type, but the broad breakdown is:
40% — Core skills and certifications match (hard requirements vs. your profile) 25% — Experience level alignment (years + seniority) 20% — Location and work authorisation compatibility 15% — Sector-specific signals (industry, company size, specialisation depth)
A 94% match means you satisfy 94% of the weighted signals for that role. It doesn't guarantee an interview — that's the recruiter's call — but it means you're a strong technical fit on paper.
What improves your score
The single biggest improvement comes from specificity in your profile:
- List every certification by its full official name, not abbreviations
- Include exact employment dates — the AI can't infer experience without them
- Add a skills section with both soft and technical skills — these are weighted
- For healthcare: list your specialisation wards explicitly (ICU, HDU, NICU, etc.)
- For engineers: list the specific software and standards you've worked with (AutoCAD, Revit, ASME, ISO 9001, etc.)
- Keep your profile updated — stale profiles score lower for recency-sensitive roles
What it doesn't do
The AI doesn't make hiring decisions — it ranks and surfaces. Every shortlist is reviewed by a human recruiter before candidates are contacted. The system also doesn't discriminate by name, nationality, gender, or age — these fields are not used as matching signals.
Match scores are a starting point for the recruiter conversation, not a gate. A 70% match who writes a compelling cover note will outperform a 92% match who submits a bare CV every time.
Ready to take the next step?
Use EmpireO's tools and live job listings to move forward.
Try AI matching now