Surface the right candidate for the right role, automatically
AI matching ranks incoming applicants and rediscovers passive candidates from your talent pool against every open requisition — updated as soon as a new job or applicant lands.
Everything you need to run this end-to-end
- Continuous scoring across all requisitions
- Silver-medallist and passive-candidate rediscovery
- Recruiter-tunable weightings (skills, tenure, location, language)
- Emiratisation-aware ranking where enabled
Screening asks: how good is this candidate for this job? Matching asks a bigger question: which candidate — anywhere in our history — is the best fit for the job we just opened? Nukhba's AI matching engine answers that second question continuously, for every open requisition, against every candidate in your tenant.
The moment a new req is approved, the matching engine scores it against your entire talent pool — active applicants, silver medallists from past reqs, passive candidates you sourced last quarter, alumni, event contacts. The top matches surface in the recruiter's queue before the job is even posted externally. The first hire on the req is often someone you already knew.
Recruiters can tune the weightings — skills, tenure, location, language, salary band, Emiratisation eligibility — and the ranking updates live. Every weight change is versioned, and the model explains in plain language why a candidate moved up or down. It is not magic; it is grounded, explainable retrieval that a hiring manager can challenge.
What's actually in the box
Continuous cross-req scoring
Every candidate is scored against every open req; every new req is scored against your entire talent pool — updated in real time.
Silver-medallist rediscovery
Strong candidates who didn't land a past role are surfaced automatically for new openings that match their profile, with the past evaluation attached.
Recruiter-tunable weights
Adjust skills, tenure, location, language, salary, Emiratisation and custom-criteria weights in one click — see the ranked list update instantly.
Explainable delta on every rank change
When weights change, the model reports which candidates moved and why — in plain bilingual language, with evidence.
Passive-candidate signals
Opportunity Network opt-ins, career-site returning visitors, and portal activity feed the matching engine — the pool is not just past applicants.
Match-quality feedback loop
Hiring outcomes and panel scores feed back into the rubric tuning suggestions — the engine learns what a good hire looks like on your reqs, in your tenant only.
A conglomerate fills 60% of new reqs from its own pool
After 18 months on Nukhba, 60% of the group's new hires now come from candidates who first applied for a different role. The saving in external sourcing fees more than pays for the platform, and time-to-hire on rediscovered candidates is under two weeks because the panel has the past evaluation on file.
Three steps, always in this order
Score everyone, always
New candidates are scored against every open req; new reqs are scored against your entire talent pool.
Rediscover silver medallists
Strong candidates from past reqs who didn't land the role are surfaced automatically for new openings.
Tune and re-rank
Recruiters adjust rubric weights in one click and see the ranking update — every version stored for audit.
What buyers ask most
Ready to modernise your hiring?
Book a 30-minute walkthrough with a UAE recruitment specialist.