TARUNAI Systems Engineer← All work
01 DOCUMENT INTELLIGENCE · CASE STUDY

Turning physical cash vouchers into validated accounting records.

Accounts AI connects scanning, OCR extraction, human validation, approval, history tracking, and Oracle storage in one controlled workflow.

Role
Primary solo ownership
Environment
Internal accounting workflow
Stack
Python · Flask · OCR · Oracle
30+Scanned voucher files used in development and testing
50–70%Estimated processing-time reduction per voucher
1–2 minEstimated reviewed workflow after extraction

Time reductions are workflow-based estimates from development testing. Formal organisation-wide savings have not been measured publicly.

Manual entry made every voucher a small, repetitive data task.

Staff had to read each physical cash voucher, identify multiple fields, enter the information, verify it, and preserve an accurate record history.

The challenge was not simply reading text. Documents vary in clarity and layout, so uncertain extraction needed to remain visible and correctable before any accounting record was approved.

Automation accelerates entry. People retain control.

The workflow separates machine extraction from accountable approval.

  1. 01

    Scan

    A physical voucher enters through the scanner workflow.

  2. 02

    Extract

    OCR identifies text and maps accounting fields.

  3. 03

    Review

    A person checks, corrects, and completes the record.

  4. 04

    Approve

    Only reviewed information can move forward.

  5. 05

    Store

    The approved record and history are stored in Oracle.

One review surface for the document and its structured record.

The interface shown here uses fictional values. No client documents, identities, or production data are displayed.

Reliability comes from the system around OCR.

01

Structured extraction

Detected text is mapped into specific accounting fields instead of accepted as an unstructured block.

02

Human validation

Users can review and correct uncertain fields before approval.

03

Approval boundary

Extraction alone does not authorize a final accounting record.

04

History and traceability

Processed documents and status changes remain accessible for later review.

Primary independent engineering across the complete workflow.

I built the platform primarily independently, while domain users helped explain requirements and validate business correctness.

01Workflow discovery and system architecture
02Flask backend and OCR processing pipeline
03Structured extraction and validation interface
04Cash-voucher entry and approval workflow
05Oracle integration and document history
06Windows scanner agent and deployment debugging

What this evidence does—and does not—claim.

Have a document workflow worth automating?

I’m available for selected document-intelligence and business-automation projects.

Discuss a project ↗Back to portfolio