![[FREE EBOOK] Strategic IT Outsourcing: Optimizing Cost & Workforce Efficiency](https://vti.com.vn/wp-content/uploads/2023/08/cover-mockup_ebook-it-outsourcing-20230331111004-ynxdn-1.png)
Our customer is a bank in Japan, with a legacy system built on VB.NET that has been running for nearly two decades, and required a comprehensive upgrade to meet modern technological standards. This legacy system was running on Windows Server 2003 with outdated database structures, making it incompatible with the PHP 8 Framework, the chosen technology for the migration.
Additionally, a significant challenge was the need to migrate the logic of handheld devices from the old platform to Android while integrating their functionality into the new web system. This task was complicated by the documentation, making VTI’s team rely heavily on analyzing the legacy source code to understand system logic.
Implementation Approach
Reconstructing the Old System
VTI’s team began by reconstructing the customer’s original development environment to ensure a smooth transition:
- Configuring Windows Server 2003 to match the legacy system.
- Analyzing and interpreting the legacy source code to understand system functionality.
Execution Process
The project was divided into two parallel teams to streamline workflows:
- Team 1: Focused on analyzing the legacy source code and drafting a comprehensive Software Requirements Specification (SRS). This included designing a new system structure and detailing the complex logic extracted from the original code.
- Team 2: Developed the new system based on the SRS, converting the system to PHP 8 Framework, and designing additional features.
This division of labor enabled our team to meet milestone deadlines effectively and minimize task overlap between groups.
Cutting 1-2 Days of Coding Down to 2-3 Hours
AI, with its natural language processing and code analysis capabilities, became a central tool in this project. VTI created custom prompts to utilize ChatGPT effectively in the following areas:
- Documentation Support: For source code files with thousands of lines, AI helped generate flow diagrams from the legacy code, visually illustrating system logic. This enabled the team to understand the old system’s structure better and ensured accuracy during the redesign process.
- Coding Assistance: AI-generated code automatically, reducing development time from several days to just a few hours.
The use of AI in code generation helps reduce the time from 1-2 days to just 2-3 hours.
AI not only supported developers but also helped testers and other team members quickly understand system logic through accurate flow diagrams and well-documented materials.
Harnessing AI to Enhance Productivity at VTI
The real power of AI came in the form of improved accuracy and speed, ensuring that the migration process moved forward without unnecessary delays. What would have taken weeks of manual work was accomplished in mere hours, giving the team a massive boost in both time and output. This application of AI has allowed VTI to deliver high-quality results faster, enabling our teams to focus on higher-value tasks while ensuring the accuracy, clarity, and efficiency needed to meet client expectations and tight deadlines.
>> Read more: VTI Pioneers the Use of Generative AI to Enhance Productivity
The success of this project further reinforces how AI can elevate the productivity and capabilities of our teams, making us more agile and innovative in the face of modern technological challenges. Contact us to discover how we can help you boost productivity.
