KreditBeeDvar

KYC and income documents

Document reading on a small model, with checks it cannot skip

One harness for every document a loan application carries: PAN, masked Aadhaar, the cancelled cheque for the e-NACH mandate, a bank statement and a salary slip. A small model reads first, a large model only for the hard field, and format, arithmetic and cross-document checks decide what goes to a person.

INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT INCOME TAX DEPARTMENT आयकर विभागINCOME TAX DEPARTMENTemblemomittedभारत सरकारGOVT. OF INDIAस्थायी लेखा संख्या कार्डPermanent Account Number CardABQPS7712Kphotopattern, no dataनाम / NamePRIYA RAGHAVAN SHARMAपिता का नाम / Father's NameRAGHAVAN K SHARMAजन्म की तारीख / Date of Birth14/08/1993हस्ताक्षर / SignatureSPECIMENSPECIMEN · NOT A VALID DOCUMENT · FICTITIOUS PERSON · DVAR DEMO

Identity and PAN check. Used in: Loan application. 1 page. Specimen drawn to the real layout, for a fictitious person and bank; the emblem, logos and real QR data are left out.

  1. 1. Classify
  2. 2. Read fields
  3. 3. Check formats
  4. 4. Cross-check
  5. 5. Route
  6. 6. Decide
Press Run the harness to read this document field by field.

Review queue

Fields under the 80% bar after routing, or named by a failed check. Each decision is saved as eval and training data.

Run a document to fill the queue.

This batch

Documents with no manual touch

–

Field accuracy vs answer key

–

Reviewer decisions saved

0

Cost vs all-large

–

Measured live from the runs above, against the answer key we wrote for these 5 documents. Large-model price assumed at $0.012 a page.

Schemas first

Each document type has a field schema; the model fills it and the schema rejects anything malformed.

Checks a model can't skip

PAN and IFSC formats, real dates, totals that add up, names that match across documents.

Routing on cost

A small vision model on Dvar first, a large model only for the hard page, cost tracked per document.

Review that teaches

Low-confidence or failed fields go to a person; their corrections become eval and training data.

Traces and evals

Every run traced in Dvar Evals with an eval set per document type, so a prompt or model change is tested first.

Data handling

PII masked in logs, India hosting, retention rules that fit DPDP.

Written for the demo: the 5 specimen documents (drawn to the real PAN, masked Aadhaar, CTS-2010 cheque, bank statement and salary slip layouts, for a fictitious person and bank) and what each model read from them, with its confidence. Real code running here: the schemas, the checks, the routing rule, the review queue and the scoring. The next step is to run this harness on a redacted sample of KreditBee's own documents.