Platform · KYC
Identity document authentication, biometric face match, and liveness detection returned as a single decision with the evidence attached to a case record your auditor can reopen years later.
curl -X POST https://api.verifix.ae/v1/verifications \
-H "Authorization: Bearer $VERIFIX_KEY" \
-d '{
"type": "kyc",
"journey": "uae_vasp_retail",
"document": { "type": "emirates_id", "front": "...", "back": "..." },
"selfie": "..."
}'
{
"id": "ver_8H2K...",
"decision": "clear",
"risk_score": 18,
"checks": {
"document_authenticity": "pass",
"biometric_match": 0.94,
"liveness": "pass",
"sanctions": "no_match"
}
}What it checks
No bundled black box. Each check is listed, configurable, and visible in the decision it produced.
Emirates ID, passport, and residency permit checks with tamper, font, MRZ, and security-feature analysis. Coverage list: UAE-issued documents today — Emirates ID, UAE passports, and residency documents.
Selfie-to-document comparison returning a similarity score, not a yes/no — so you set the threshold your risk policy requires.
Passive and active liveness to defend against printed photos, screen replays, and deepfake injection. Presentation-attack testing: presentation-attack detection is applied on every liveness capture.
Name, date of birth, nationality, document number, and expiry extracted and normalized, ready to pre-fill your onboarding form.
The extracted identity is screened against sanctions, PEP, and adverse-media sources before the decision returns — no second vendor round trip.
Periodic refresh triggered by document expiry, risk-score change, or your own review cadence, with the prior evidence kept intact.
How it runs
The same sequence whether you call the API directly or use the hosted flow.
Call the API with a journey ID, or drop in the hosted flow and let VerifiX collect the document and selfie on your behalf.
Blur, glare, crop, and framing are checked at capture so a customer fixes the photo instead of failing a check minutes later.
Document authenticity, biometrics, liveness, and screening run in parallel and feed one weighted risk score.
Clear, review, or blocked. Anything not automatically cleared lands in the review queue with every artefact attached.
Scan your Emirates ID
Place the front of the card inside the frame
Illustrative hosted flow — no customer data shown
Modules
Each one is a separate module you can switch on, order, threshold, or skip inside a journey.
Prove the person is who they claim to be, that the document is genuine, and that the data on it is usable downstream.
document_verificationAuthenticates an identity document: MRZ parsing and checksums, font and layout comparison against the issuer template, security features, tamper and digital-edit detection, expiry.
nfc_chip_readReads the contactless chip in an e-passport or Emirates ID and validates the issuer's digital signature, so identity data comes from the issuing authority rather than from an image.
data_extractionExtracts name, date of birth, nationality, document number, issue and expiry dates, then normalizes transliteration (including Arabic name handling) so records match across systems.
database_validationConfirms the extracted identity against authoritative and bureau data sources rather than trusting the document alone.
proof_of_addressAccepts a utility bill, bank statement, tenancy contract, or Ejari document, extracts the address, checks document age, and compares it against the declared address.
phone_email_verificationOne-time passcode ownership check plus reputation signals on the number and mailbox (disposable domains, recently created, carrier and line type).
questionnaireCollects the declarations your regulator expects — source of funds, source of wealth, purpose of relationship, expected volumes, tax residency — inside the same session as the identity checks.
Prove the person is physically present and not a replay, a mask, a deepfake, or an account already rejected under another name.
passive_livenessConfirms a live human is in front of the camera from a single capture, with no instructions to follow — defeats printed photos, screen replays, and video injection.
active_livenessChallenge-response capture (movement or on-screen prompts) for high-risk journeys or where a regulator expects an explicit challenge.
face_matchCompares the selfie against the document portrait and returns a similarity score, not a yes/no, so your policy sets the accept, review, and reject thresholds.
face_searchSearches the new face against your own verified population and your blocklist to catch one person opening several accounts under different identities.
biometric_authRe-proves an existing customer at a sensitive moment — large withdrawal, device change, payout address change — by matching against the biometric captured at onboarding.
age_estimationEstimates age from a face capture where a full document check is disproportionate, with document fallback when the estimate lands near the threshold.
device_ipFingerprints the device and inspects the network path: emulator, rooted device, automation, VPN, proxy, hosting ASN, geo mismatch against the claimed residency.
duplicate_blocklistBlocks identities, faces, documents, devices, and wallets you have already rejected, and flags duplicate applications inside the same tenant.
See the full module catalog — every KYC, KYB, AML, KYT, and platform module with its inputs and outputs.
Evidence retained
Questions
A 30-minute walkthrough with a compliance engineer, or a sandbox key in your inbox today.