Skip to content
← All selected work

SimpleBooks (Pvt) Ltd. / Case study

An AI tax consultant that reads your paperwork

What changes when the paperwork is in the languages people actually use?

Contribution
Software Engineer (AI & Full-stack)
Period
Apr 2025 — Mar 2026
Location
Colombo, Sri Lanka
Sinhala + Tamil

NIC text extraction

Payslips + T-10

Structured tax inputs

Custom API

Registration integration

Reported in the supplied CV. See the evidence note below for scope and measurement limitations.

01 / The situation

Tax prep and company registration meant slow, manual data entry from payslips, T-10 forms and national IDs — with no room for the two other national languages people actually use.

At SimpleBooks, the work covered an AI tax consultant, payslip and T-10 extraction, and national identity-card processing for company registration.

What made it an integration problem

The system had to connect Sinhala and Tamil text to structured director details, then deliver those details through a custom API and frontend integration.

02 / My contribution

The parts I built.

  1. 01

    Developed an AI multi-agent system that acts as an intelligent tax consultant.

  2. 02

    Built a tax tool that automatically extracts and structures data from payslips and T-10 forms, eliminating manual entry.

  3. 03

    Shipped a multilingual NIC OCR system that reads Sinhala / Tamil IDs and auto-populates director details in a business-registration platform via a custom API and frontend.

Multi-agentOCR (Sinhala/Tamil)Custom APIFull-stack

03 / The system

Follow the information.

  1. 01NIC image · Sinhala / Tamil
  2. 02Multilingual OCR
  3. 03Structured director details
  4. 04Custom API
  5. 05Registration frontend

Conceptual workflow reconstructed from the CV. This is not a screenshot of a client system.

04 / An engineering reading

Why the connections matter.

The deliverable included the API and destination interface alongside OCR. Extracted information could become registration data rather than another document for someone to copy from.

The tradeoff to pay attention to

Multilingual support expands who a workflow serves. It also requires evaluation across scripts and document conditions. The CV establishes the supported scripts and integration, but does not specify dataset size, accuracy by script or exception-handling rules.

This section interprets the delivered scope. It does not claim an undocumented decision process.

05 / What supports this story

Keep the claims close to the evidence.

The CV confirms automatic extraction from payslips and T-10 forms, a multi-agent tax consultant, and Sinhala/Tamil NIC extraction integrated into business registration. No numerical accuracy or time-saving benchmark was supplied.

Read the source CV ↓

Another problem, another system

From a conversation to an order.

Discuss a similar problem ↗