About the Company
The company is a technology provider that develops digital solutions to support organizations and individuals across various workforce and employee needs. Its product ecosystem covers areas such as human resources and payroll management, employee attendance and productivity, employee financial benefits, training and development, and career services.
With multiple solutions serving different aspects of the employee lifecycle, the company continuously develops and maintains applications across its product portfolio. This requires its technology environment to evolve alongside changing business needs, new product requirements, and technology standards.
As the applications grew over time, keeping the underlying technology up to date became increasingly important. The company needed to modernize parts of its existing applications while maintaining security, performance, and long-term maintainability. This led the team to explore a more efficient approach to application modernization and development.
Challenge: Modernizing Existing Applications
As part of its application modernization initiative, the company identified several areas of its existing applications that required significant updates. The modernization involved multiple technology upgrades and architectural changes, including upgrading PHP version from 7.2 to 8.4, updating Next.js, migrating from Ant Design and SCSS to Shadcn UI and Tailwind, moving from Pages Router to App Router, replacing Redux with Zustand and TanStack Query, and implementing TypeScript.
With multiple dependencies and components involved, performing the rewrite manually would require significant development effort and time.
To address this challenge, the company began exploring AI coding tools to support the rewrite process. The team found that working with individual AI tools in isolation meant constantly switching between different tools for tasks such as writing code, reviewing architecture, and maintaining documentation. When the team discovered Kiro, they found a single environment where AI assistance is integrated directly into the development workflow, from requirements and design through to implementation. Kiro also provides access to multiple AI models from different providers, including OpenAI, Anthropic, DeepSeek, and others. This gave the team greater flexibility to select models based on task complexity, while managing credit consumption according to their development needs.
Solution: Accelerating Application Modernization with Kiro
To accelerate the modernization effort, the company worked with CDT to explore how Kiro could be integrated into its existing development workflow. CDT conducted a hands-on workshop based on the company’s modernization needs, demonstrating how Kiro’s Spec-Driven Development could help structure and support the application rewrite process. In this approach, development begins with a structured “spec”, a document that captures requirements, design decisions, and task breakdowns before any code is written. This gave the entire team, from engineers to tech leads, a shared understanding of the modernization plan upfront, reducing the risk of misalignment mid-sprint. It also meant the development process was documented automatically within Kiro.
During the workshop, the CDT team worked with the company’s developers to walk through their existing development challenges and demonstrate how Kiro could be applied to the rewrite process. The team explored Kiro’s capabilities using the company’s actual development requirements, giving the developers a clearer understanding of how AI-assisted development could support their modernization work.
Following the workshop, the company received a two-month trial to apply Kiro directly to its development environment. The team used Kiro to rewrite existing applications, implement new features, and support ongoing product development, allowing them to evaluate its capabilities through real development tasks. Over time, Kiro became part of their broader development workflow, supporting both application modernization and new feature development.
How Kiro Supported Company Modernization Journey
As the team worked on the application to rewrite, Kiro helped make the AI-assisted development process more flexible and consistent. Its multiple AI models, flexible credit usage, and built-in development features supported different needs throughout the modernization process.
Choosing the Right Model for Each Task
Application modernization involves different types of development tasks, from updating existing code and migrating frameworks to building new components. Kiro provides access to multiple AI models in one development environment, allowing developers to choose the model that best fits each task.
This gave the team flexibility to work on different parts of the application without switching between separate AI tools.
Managing AI Usage
Different AI models in Kiro use credits at different rates. This allowed the team to manage AI usage based on the complexity of each task.
For simpler tasks, developers could use models with lower credit consumption, while more complex tasks could use higher-performance models. This gave the team more flexibility in managing AI usage throughout the rewrite.
Keeping the Rewrite Consistent
With multiple developers working across different parts of the codebase, maintaining consistency can be challenging. Kiro’s Spec-driven development and Steering features helped the team establish a more consistent approach.
Spec-driven development allowed the team to define requirements, design decisions, and tasks before implementation, keeping the development process structured and documented.
Meanwhile, Steering helped define coding standards and architectural guidelines, such as naming conventions, component structures, and preferred libraries. These guidelines could then be applied consistently across different parts of the application.
Results: Faster Application Modernization with Fewer Resources
Kiro helped the company accelerate its application modernization effort while reducing the development resources required for the rewrite.
The modernization effort was initially estimated to require a team of approximately 5–6 developers working for 6 months if carried out manually. With Kiro supporting the development process, the company completed the rewrite with 2 developers in approximately 3months, reducing the estimated development timeline by 50% and the number of developers required by approximately 60–67%.
This meant the team was able to carry out the modernization with fewer developers and within approximately half of the initially estimated timeline. Beyond accelerating the rewrite itself, the experience also demonstrated how AI-assisted development could be incorporated into the team’s day-to-day workflow.
Following the modernization effort, the team continued using Kiro for product updates, new feature development, and ongoing application development. What began as a solution for a specific rewrite initiative therefore became part of the team’s broader approach to application development.
Beyond application modernization, the company is exploring additional AI use cases across its HR product ecosystem. These include a centralized sales dashboard powered by Amazon Quick, which would help the sales team analyze customer purchases, identify customer needs, and uncover opportunities to recommend additional products.
The company is also exploring ways to connect its existing databases with analytics and AI capabilities to support payment reminders and other customer-related insights. Together, these initiatives reflect the company’s continued interest in applying AI across both application development and broader business operations.
Explore AI-Assisted Development with CDT
Looking to accelerate application modernization or explore how AI can support your development workflow? Talk to Central Data Technology (CDT), an AWS Premier Tier Services Partner in Indonesia, to discover how AWS solutions can help address your business and technology needs.
Explore what’s possible with CDT and AWS here!
