Context
Face matching across a very large customer base to catch duplicate and impersonated identities, and automated classification of KYC files at a microfinance lender.
Challenge
Duplicate identities and manual KYC classification slowed onboarding.
What we deployed
Face match across the customer base and automated KYC file classification.
Typical architecture
- STEP 01
Image captured at onboarding
- STEP 02
Face embedding compared against 80M+ records
- STEP 03
Possible duplicates queued for review
Illustrative of how this solution is typically structured and governed; deployment specifics are available under NDA.
Results
- 80M+ customer records face-matchedPartner-reported
- 94% match accuracy in <40 msPartner-reported
- 330k microfinance KYC files classified automaticallyPartner-reported
NoteResults from deployments by GoGenesis and its platform partners. Outcomes vary by portfolio and operating environment; your results are set in the pilot.
Governance approach
- Matches are recommendations; staff confirm
- Biometric data handled under agreed residency rules
Illustrative of how this solution is typically structured and governed; deployment specifics are available under NDA.