LG CNS has introduced an Agentic AI-based development platform designed to support the entire process of building and operating large-scale IT systems.
On June 8, LG CNS announced the launch of its Agentic AI Development (AIND) platform.
While recent advancements in coding through natural language, known as "vibe coding," have gained attention, concerns have been raised about the potential for code generation to clash with existing systems due to a lack of understanding of enterprise system architecture. To address this issue, LG CNS has implemented a specialized AI agent tailored for the entire system development process.
AIND was developed based on extensive experience in system construction and operation across various industries. When users input requirements in natural language, the platform facilitates end-to-end development through the collaboration of multiple agents: a requirements analysis and design agent, a coding agent, and a testing and quality assurance agent.
For instance, if a financial institution requests the addition of a new financial service to its existing core banking system, such as "please build an automatic transfer service for savings accounts linked to the account system," the analysis and design agent will assess the requirements and design the system architecture. Subsequently, the coding agent will generate code in accordance with the financial institution's development standards, and the testing and quality assurance agent will validate the output. This allows users to focus on reviewing and approving results, thereby shortening the development timeline.
AIND's core competitive advantage lies in its "knowledge foundation," which integrates and analyzes IT information within the enterprise. This knowledge foundation is an ontology-based database structured to help AI understand development standards, security regulations, system source code, and various development deliverables. AIND utilizes this foundation to learn about the enterprise's systems and work environment, enabling customized development.
LG CNS co-developed AIND with Cline, a U.S.-based open-source AI coding company. Cline's AI coding agent has gained attention for its rapid growth on the global development platform GitHub.
The two companies plan to expand the application of AIND in IT system construction and operation projects in the finance, public, manufacturing, and defense sectors, focusing on the United States, Japan, and Southeast Asia.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.