Artificial intelligence is rapidly transforming software development, and LG CNS is moving to accelerate that shift with the launch of a next-generation AI development platform designed for enterprise environments. The company aims to help organizations automate the entire software development lifecycle, from system design and coding to testing and quality assurance.
LG CNS announced on June 8 the launch of “DevOn Agentic AI Native Development (AIND),” an agentic AI-based software development platform. Unlike conventional AI coding tools that primarily focus on code generation, AIND is designed specifically for large-scale enterprise system development and operations.
The launch comes as “vibe coding,” a development approach that uses natural language prompts to generate code through generative AI, gains popularity across the software industry. However, many organizations have struggled to apply these tools in complex enterprise environments where security requirements, development standards, and legacy system architectures must be carefully considered.
To address these challenges, LG CNS developed specialized AI agents tailored to different stages of software development. When users submit requirements in natural language, AI agents responsible for analysis and design, coding, testing, and validation collaborate to automate the system development process.
For example, if a financial institution wants to launch a new deposit service, users can simply describe the requirements. The AI platform analyzes the business needs, designs the system architecture, generates code that complies with development standards, and performs validation processes. Developers can then focus on reviewing and approving the results rather than performing repetitive coding tasks.
A key differentiator of AIND is its “Knowledge Foundation,” a structured framework that enables AI to understand an organization’s internal IT environment. The system integrates development rules, security policies, existing source code, and technical documentation into a unified knowledge base, allowing AI to learn and operate within enterprise-specific environments.
As a result, the platform goes beyond simple code generation and can perform customized development tasks based on an understanding of each company’s business processes and system architecture.
LG CNS has also incorporated a Spec-Driven Development approach to improve reliability and consistency. By requiring AI agents to follow predefined rules and standards during design, development, and validation, the platform reduces variations in output quality and minimizes the risk of errors.
The platform also includes legacy modernization capabilities. It can automatically convert COBOL-based systems into Java environments and restructure existing Java applications to align with modern software architectures.
Tasks that previously required weeks or even months of manual analysis, code conversion, and verification can now be completed significantly faster through AI automation. LG CNS is currently applying these capabilities to next-generation system development projects at major financial institutions in Korea.
The company is also preparing to expand globally. LG CNS developed the platform in collaboration with Cline, a U.S.-based AI coding company, and plans to target industries where security, reliability, and compliance are critical, including finance, government, manufacturing, and defense.
The strategy aligns with growing demand for AI-driven software development automation in markets such as the United States, Japan, and Southeast Asia. LG CNS aims to strengthen its position in these regions by providing enterprise-grade AI development solutions.
Looking ahead, the company emphasizes that AIND is designed to enhance developer productivity rather than replace human engineers. Repetitive development tasks can be handled by AI, while developers focus on architecture, decision-making, and innovation.
LG CNS expects the platform to help create a new software development environment where AI and human developers work together to deliver faster, more efficient, and higher-quality enterprise applications.
