Press Release
August 13, 2026
Nota AI (KOSDAQ: 486990, CEO Myungsu Chae), a company specializing in AI model optimization and compression technology, announced today that it is expanding beyond model-level optimization into the enterprise AI agent operations infrastructure market, marking its entry into AI-driven transformation (AX) of the customer service industry.
The company said it has signed a co-development agreement with MetaM, a leading AI contact center (AICC) provider, to build a next-generation AI-based customer consultation system. The agreement represents Nota AI's first commercial project in the enterprise AI agent operations infrastructure market.
This marks the first instance of Nota AI applying its optimization technology to AI operations infrastructure designed to run multiple AI agents reliably in real-world enterprise environments. Building on this initial project, Nota AI plans to expand its reach beyond customer service into other industries adopting AI agents, including manufacturing, finance, and the public sector. Through this expansion, the company aims to help enterprises operate AI services reliably and cost-effectively, even within limited AI infrastructure.
As enterprise adoption of AI agents accelerates, AI contact centers (AICC) which apply AI across the customer service process have emerged as a leading commercialization market for enterprise AI. Because AICC applications span customer response, consultation summarization, and quality management, they offer the potential to simultaneously reduce operating costs and improve customer experience, with strong potential to expand into industries such as finance, telecommunications, retail, and the public sector.
As enterprises deploy AI agents across a growing range of tasks, the resulting increase in model calls and computational load drives up GPU costs and memory usage. This is especially critical in large-scale customer service environments, where optimizing computational efficiency, response speed, and memory usage for real-world service conditions is as important as model performance itself. Nota AI is applying its proprietary optimization technology to this project to help reduce that operational burden.
Under the agreement, Nota AI will combine its AI model compression and optimization technology and AI agent design capabilities with MetaM's expertise in AI contact center operations to build an AI-native multi-agent consultation system. The system will divide consultation tasks — including customer intent analysis, information retrieval, response, consultation summarization, knowledge management, and quality control — among role-specific AI agents that work in collaboration with professional customer service agents.
As part of the project, Nota AI will develop the underlying technologies that power the system's multiple AI agents, including a shared AI engine, orchestration, and retrieval-augmented generation (RAG). The company will apply its proprietary compression and optimization technology to the large language models and inference engines used in the system, aiming to reduce computational load and memory usage while minimizing any loss in model performance. This is designed to maintain stable response speeds even when multiple AI agents are operating simultaneously, enabling cost-efficient operation in real-world customer service environments.
MetaM, Nota AI's partner on the project, is a business process outsourcing (BPO) company with more than 20 years of experience operating contact centers. The company has accumulated operational expertise running contact centers across industries including finance, retail, manufacturing, and online platforms. With a nationwide contact center infrastructure serving more than 200 client companies, MetaM has an established business foundation for deploying AI agents in large-scale, real-world customer service operations.
On this project, MetaM will contribute domain expertise, support the implementation of domain-specific functions for each role-based AI agent, and oversee the deployment and operation of the AICC system.
"In the era of AI agents, competitiveness depends not only on the performance of the AI model itself, but on how quickly, reliably, and cost-effectively it can be operated in real-world services," said Myungsu Chae, CEO of Nota AI. "This project marks an important milestone in extending Nota AI's AI model optimization technology beyond individual models into the infrastructure that operates multiple AI agents."
"We chose Nota AI's AI model optimization technology to reliably operate AI agents in large-scale, real-world consultation environments," said Eun-taek Lim, CEO of MetaM. "Together with Nota AI, we will address the core challenges of AI agent operations — inference cost and memory usage — while introducing a next-generation consultation service that improves both operational efficiency and customer experience."
