Back
Key Takeaway
Establishing standards and providing education for AI Code Assistant adoption to improve development productivity by 30%
Through AIR DevOps, we selected the optimal AI Code Assistant and established usage standards. This achieved a 30% improvement in developer coding productivity and a 15% reduction in code merge time by reducing repetitive tasks, successfully internalizing an AI-based development culture.
Airlines (H Company)
Client :Airlines (H Company)
Industry :Data & AI / Airlines / Transportation
Service Area :Applications & DevOps / Data & AI
Applied Solution :AIR
1. Overview (Project Background)
H Company pursued a new development project for an employee portal groupware system to innovate work efficiency and user experience (UX) as a key strategy for accelerating digital transformation (DX). This project went beyond simple system construction and was an important task that needed to secure both development speed and quality while laying the foundation for future systems.
H Company adopted the AIR DevOps methodology to accelerate and successfully complete this core system development. In particular, by systematically introducing AI Code Assistant technology into the internal development environment, the company set goals to standardize AI utilization capabilities across the entire development organization and internalize AI-based development methodology (AI Driven Development) as key to the success of the groupware project and dramatic improvement in development productivity.
2. Challenge (Problem Definition)
Inefficient requirements analysis and planning:
In the early stages of groupware development, collecting vast employee requirements and manually analyzing and planning them according to the latest UX/UI trends consumed significant time and resources, resulting in inefficiencies in the initial design phase.
Confusion in AI tool adoption and lack of standardization:
With various AI Code Assistants available in the market, there were difficulties in selecting the optimal tool and standardizing its usage methods.
Disparities in AI utilization capabilities and repetitive tasks:
AI Code Assistant utilization varied depending on individual developer capabilities, and resources were invested in repetitive development tasks such as application coding and error fixing, making it difficult to focus on creative work.
3. Solution (Resolution Approach)
Megazone Cloud provided an End-to-End solution based on the AIR DevOps methodology, combining systematic consulting based on PMP and AI Driven Development based on Agile.
AI-based requirements analysis and design acceleration: By applying AIR's AI-Driven SDLC, we rapidly analyzed natural language-based requirements and trend data, dramatically shortening the initial planning phase by converting them into detailed groupware specifications. (Early application of AI value in the consulting phase)
AI Code Assistant standardization: We functionally validated various AI Code Assistant technologies suited to the groupware development environment and selected tools optimized for the customer environment (Claude, Amazon Q Developer, etc.) to establish development environment standards.
Systematic training and practical application: To enhance developer capabilities, we developed AI Prompt writing guides and Frontend/Backend development guides, conducted systematic hands-on training, and applied AI-based code generation and error fixing processes to actual development.
4. Result (Achievements)
Through the adoption of AIR DevOps and standardization and internalization training of AI Code Assistant, we achieved the following tangible results.
30% improvement in code productivity
15% reduction in code merge time
Improved development focus through reduction of repetitive tasks and establishment of AI-Driven SDLC process
Upward leveling of capabilities across the entire development organization through internalization of AI utilization standards






