Data Extractor

As the lead developer for the Data Extractor project, I built a powerful Kotlin API that automates the extraction and processing of data from a wide range of documents. This project has streamlined the handling of tens of thousands of documents, saving our team hundreds of hours of manual data processing.

Project Overview:

The Data Extractor API accepts a file and its associated document type (e.g., AFS, Bank Statement, Cheque, General Information Sheet, Income Tax Return, Security Certificate). It then automates the process of extracting critical data from these documents and converting it into a standardised JSON schema format required by our internal systems.

• Document Parsing: The project leveraged Azure AI Form Recognizer to extract structured and unstructured data from documents with high accuracy.

• Data Mapping: The extracted data was processed using the OpenAI API’s Chat Completions feature to map the information into a well-defined JSON schema format. This allowed seamless integration with our databases.

• Automation & Efficiency: The automation of this document workflow eliminated the need for manual data entry and significantly reduced processing time, enhancing overall productivity.

• Backend: Built entirely in Kotlin, the API utilized the robust and reusable Kotlin Base project I had developed previously.

• Azure AI Form Recognizer: Handled the initial data extraction phase.

• OpenAI API: Used Chat Completions to transform raw data into a structured JSON schema, ensuring the information was accurately and consistently stored in our databases.

Impact:

This project not only increased efficiency but also reduced the potential for human error in data processing. It enabled our team to focus on higher-value tasks while automating repetitive document handling, contributing directly to operational scalability and business growth.