New platform lets businesses choose only the AI features they need

[Financial News] Megazone Cloud announced on the 6th that it had launched AIR Studio v3, an enterprise AI platform that lets companies select and apply only the features they need, depending on their level of AI adoption. It is designed to integrate management of data, AI agents, operations, security and costs while maintaining existing AI environments.
A key feature of AIR Studio v3 is that it divides core functions into four areas: Data, Agent, Ops and Governance. Rather than adopting the entire platform at once, companies can start with the functions they need for immediate priorities, such as data preparation, agent development, or security and cost controls, and then expand in stages.
Megazone Cloud has incorporated into this modular structure the experience it gained deploying the platform at domestic and international companies in finance, retail, manufacturing and construction after launching the first version in August last year and the second in April this year.
The Data module automates the collection, processing and storage of company data in a format that AI can use. It is equipped with a dedicated engine that applies processing methods suited to different data types, with the aim of improving AI’s data recognition accuracy and retrieval-augmented generation (RAG) performance.
The Agent module supports the development of AI agents for business use and their connection to external tools. It comes with connectors for more than 70 business applications, including Google Workspace, Jira and Slack. Megazone Cloud said that integrations with external tools—which previously could take up to several tens of days because of the need for separate development and certification—can now be completed within minutes.
The Ops module centrally manages the performance and permissions of agents developed in AIR Studio as well as those developed externally. It also supports tool verification and integration for servers based on the Model Context Protocol (MCP), a standard for connecting AI with external tools. Administrators can monitor deployment status, performance and logs in real time, and stop or redeploy agents if necessary.
The Governance module applies common security and usage policies across various AI models, agents and external tools. Its self-developed AI Gateway centrally manages AI calls and provides features including issuing Virtual Keys to prevent exposure of original authentication keys, tracking audit logs and blocking calls that bypass controls.
The platform’s cost management features have also been enhanced. In the FinOps dashboard, token usage limits and budgets—the token being the unit of text processed by AI—can be set separately according to factors such as organization, user role and rank. Actual usage and projected costs can be monitored in real time, and calls can also be restricted if a set budget is exceeded.
James Kong, Megazone Cloud’s chief AI officer (CAIO), said, “We plan to offer industry-specific AI applications through the AIR Studio Marketplace. With close support from field deployment engineers (FDEs), we will shorten the transition process so customers can move beyond proof of concept (PoC) and generate real business results.”
[email protected] Yoon Hong-jip Reporter