AML Screening

Screen individuals against global sanctions, PEP, and watchlists.

How It Works

Valydar maintains embedded lists from:

• OFAC SDN — US Treasury sanctions
• EU Sanctions — European Union consolidated list
• UN Security Council — United Nations sanctions
• HMT — UK HM Treasury sanctions

Name matching uses Levenshtein distance and Jaccard similarity for fuzzy matching.

Running AML Screening

AML screening runs as part of the verification pipeline. Include "aml" in the checks array when creating a verification:

curl -X POST https://api.dev.valydar.com/verifications \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "client_reference": "user_123",
    "checks": ["aml"]
  }'

The name to screen is extracted from the client_reference field.

Result

{
  "passed": true,
  "details": {
    "sanctions_matches": [],
    "pep_matches": [],
    "watchlist_matches": [],
    "risk_level": "low"
  }
}

Risk Levels

Level	Meaning
low	No matches found
medium	Possible name similarity — manual review
high	Strong match — requires compliance review

Notes

• Screening is name-based only (no biometric matching)
• Fuzzy matching may produce false positives for common names
• Results should be reviewed by compliance officers for high-risk matches
• Sanctions lists are updated regularly
• The AML connector supports 100+ PEP names across 30+ countries
• Adverse media screening is included for flagged subjects
• Database-backed sanctions data can be loaded from PostgreSQL for production use