GS Neotek announced Wednesday that it successfully showcased its generative AI and security governance solutions, optimized for global technology trends, at the AWS Summit Seoul 2026 held at COEX in Seoul.
The company operated a booth at the event, where it introduced its AI agent builder platform "MISO," among other offerings. MISO is a no-code platform that allows non-experts to create and deploy business AI apps and agents through intuitive workflows without coding. It maximizes scalability by supporting integration with internal and external data based on multi-LLM integration, as well as connectivity with collaboration tools such as Teams and Slack. The platform drew high praise from corporate attendees for applying customized security policies tailored to each company to fundamentally prevent sensitive information leaks, and for enabling rapid prototype development through industry-specific preset templates.
The company also demonstrated its differentiated innovation by presenting "aiDea," a natural language-based AX agent; "NL2Device," an on-device agent that operates smartphones directly through voice commands; and "SQLens," which extracts and analyzes data through natural language queries.
GS Neotek also introduced a management platform designed to ensure stable post-deployment operations and realize "cloud infrastructure security governance." The infrastructure security governance solution "Security Lens" drew attention as a tool that automatically diagnoses cloud assets across multiple accounts and regions and visualizes actual attack paths. In conjunction with this, the company presented "Prompt Lens," an AI response evaluation and prompt diagnosis platform that quantitatively evaluates responses by LLM and provides prompt optimization reports based on multi-model comparisons, offering IT and security teams an efficient direction for governance.
Alongside the solution exhibition, a specialized presentation session by Kim Sung-hyuk, AI Research Engineer at GS Neotek's AI Center, was successfully held on the first day of the event. Kim delivered a presentation on "GPU Operation Optimization Based on DRA (Dynamic Resource Allocation) in LLM and Agent Environments," presenting a next-generation operating model that maintains AI workload performance while reducing infrastructure costs.
