# Kita Capture

## Every document. Every number. Every fraud signal — in seconds.

Pulls credit signals from any document — typed, handwritten, photo, scan — and unifies them into one structured view ready for the decision. Not raw data for your team to clean.

**Accuracy 98%**  
**Per document <30s**  
**Document types & counting 50+**  
**Used across Microfinance·SME & commercial·CDFI**

### Processing documents across the markets we serve

- Bank statements  
- E-wallet records  
- Government IDs  
- Tax documents  
- Payslips  
- Audited financials  
- Business financials  
- Credit reports  
- Utility bills  
- Mobile money records  
- Insurance documents  
- Invoices & receipts  
- Loan applications  
- Property deeds  
- Vehicle registrations

## How it works

### Pixels in. Structure out.

Kita Capture reads any document with vision language models — typed, handwritten, photo, or scan — and returns clean, schema-checked structured data. Fraud signals run alongside extraction, not after.

Extracting

```plaintext
bpi_statement_oct.pdf · page 3 of 14

Extracting

BPI

Bank of the Philippine Islands

Statement of Account

01 Oct 2025 — 31 Oct 2025

Account holder Maria Santos Reyes

Account no. 8432-1184-22

Currency PHP

Total credits ₱ 1,247,852.10

Total debits ₱ 1,809,460.88

Closing balance ₱ 640,552.77

Date          Description                                Amount            Balance
06 Oct       Salary credit · Reyes Trading Co.         + 142,500.00      492,103.55
07 Oct       GCash transfer · 09177234278            + 432.10          1,247,852.10
12 Oct       CASH DEPOSIT · branch-0142              + 1,809,460.88      1,809,460.88
15 Oct       SSS contribution                           4,250.00          1,803,210.88
21 Oct       Loan repayment · BPI Personal Loan      18,492.00         916,331.55

Structured output / bank_statement
Schema OK
```

```json
{
  "applicant": {
    "name": "Maria Santos Reyes",
    "account_no": "8432-1184-22",
    "institution": "BPI"
  },
  "period": { "start": "2025-10-01", "end": "2025-10-31" },
  "cashflow": {
    "opening":  482183.65,
    "inflow_total":  1247852.10,
    "outflow_total": 1809460.88,
    "closing":  640552.77,
    "net_recurring_credits": 142500.00
  },
  "transactions": 73,
  "confidence": 0.98
}
```

### Fraud signals

1 signal  
Font mismatch on transaction line page 2 — 6th row · low

### Statement totals reconcile
opening + credits − debits = closing

### Document metadata intact
PDF producer matches BPI fingerprint

### No duplicate of prior submitted file
perceptual hash unique vs 27 prior files

## Capabilities

### Credit signals out, not raw data.

Capture reads context, structure, and content the way a trained analyst would. Then it returns the credit signals your underwriters actually use, metrics, qualitative reads, metadata, fraud.

- Credit-signal extraction — Metrics, qualitative reads, metadata, and fraud, every kind of signal an underwriter needs, pulled from one document and ready to feed your policy engine.
- Reads like an analyst — Layout, tables, handwriting, context. Catches the things a template OCR misses, including the things written between the lines.
- Fraud detection — Tampering, metadata inconsistencies, and forgery patterns calibrated to each market and document type.
- Cross-document verification — References data points across multiple documents in the same file to catch contradictions, restating income, mismatched IDs, doctored balances.
- Multi-format support — PDFs, photos, scans, screenshots, handwritten records, processed identically. No pre-processing required.
- Market-calibrated — Tuned to local document formats across Southeast Asia, Latin America, and the United States. Not generic.

```json
{
  "document_type": "bank_statement",
  "institution": "BDO Unibank",
  "period": "Sep–Dec 2025",
  "avg_monthly_balance": 142580.50,
  "net_inflow_3mo": 485200.00,
  "dscr": 2.4,
  "income_pattern": "stable_recurring",
  "unusual_activity": false,
  "fraud_score": 0.02,
  "tampering_detected": false
}
```

## Coverage

### 50+ document types

Capture handles any document format out of the box. No templates, no pre-configuration required.

- Bank statements
- E-wallet records
- Government IDs
- Tax documents
- Payslips & income
- Business financials
- Credit reports
- Utility bills
- Mobile money
- Invoices & receipts
- Employment records
- Insurance documents

## Integration

### Deploy in minutes

- **Python SDK:** Install, authenticate, and start processing in under 10 lines. Type hints and async support included.
- **REST API:** Language-agnostic HTTP endpoints with webhook callbacks. Output in JSON, CSV, Excel, or custom. OpenAPI spec included.
- **Web portal:** No-code portal to upload documents, review extractions, and export results. No engineering required.

## Signal validation

### On request, for lenders with a book

Want to know which signals actually predict default? We'll prove it on your book.

- **Replay on your historical applications:** Hand us a sample of historical loans and we re-run Capture across the documents, pulling the full signal pack on every borrower file.
- **Match each signal to repayment outcomes:** At any DPD horizon you track. Early delinquency at DPD 1 or 4, mid-book at DPD 30 or 60, charge-off at DPD 90 and beyond. Per-signal, per-window.
- **Ranked report, what splits risk:** Each signal gets an Information Value score, monotonic decile splits, and a strength label. You see which signals belong in your scorecard and which don't move the needle.

## Try it

### Free credits in the portal

Underwrite more borrowers. Faster. With cleaner data.