Celltrion Healthcare Achieves 70% Product Completeness in 4 Days with OMS Build Workshop Using AI-DLC and Kiro

Celltrion Healthcare applied AWS's AI-DLC (AI-Driven Development Life Cycle) methodology and the AI-based IDE Kiro to resolve constraints such as a tight schedule (3 months), complex domain, and a shortage of senior developers in the OMS (Order Management System) build project for pharmaceutical CMO management and supply chain integration.

The project was conducted as a 1-day pilot and a 3-day main workshop, targeting a product scale of 5 main menus, 16 pages including submenus, and 46 DB tables. AI-DLC is a methodology that positions AI as a core collaborator throughout the entire SDLC, consisting of three stages: Inception (requirements analysis), Construction (design and code generation), and Operation.

As a result of the workshop, in the Inception stage, AI generated 98 User Stories and aligned points of disagreement between planning and business operations in real time. In the Construction stage, functional design, ERD, API specs, and code were generated sequentially, and the product completeness was evaluated at approximately 70% as of day 3. Additionally, supplementary deliverables such as design documents, architecture diagrams, and Runbooks for security audit response were secured.

Key achievements include reduced communication costs among planners, developers, and business stakeholders, improved visibility through templated deliverables, and enhanced domain understanding through direct participation by non-developers. Celltrion Healthcare concluded that AI-DLC narrowed the gaps between schedule, quality, and security compliance, and that the most significant change began with the introduction of a collaborative approach where stakeholders communicate in the same space.