South Korean startup WEVEN has launched Zaemit.ai, a professional version of its AI-assisted web-production platform aimed at agencies, designers, developers, and freelancers managing multiple client sites. The July 2026 update combines collaborative project management, reusable components, AI-generated design work, and conversational back-end development in one service.
As reported by IT Dong-A, Zaemit uses specialized agents for planning, design, coding, and back-end work. Users describe what they want in natural language, while the platform assembles a website and its supporting features rather than limiting output to a static page mock-up.
The most substantial additions are a Workspace system, Block Management, and a conversation-driven back-end implementation tool.
Workspace is designed as a central dashboard for projects, team members, site status, client communication, domains, and hosting expiration dates. That is a practical feature for agencies and freelancers that routinely juggle renewals and handoffs across several customer sites.
Block Management lets teams save and reuse common interface components or layouts. WEVEN says users can attach detailed AI-generation guidance to blocks, helping preserve a customer’s typography, spacing, colors, and component choices when creating repeat pages or new projects based on an established template.
The company is positioning the service less as an AI site generator and more as a workflow layer that stays involved after the initial page is produced. It also says Zaemit can bring in HTML created with external AI tools for editing and preparation for hosting.
WEVEN also claims that Apollon reduces token use and generation costs compared with general-purpose large language model workflows. Those are vendor performance claims, and prospective users should test output quality, editability, pricing, and usage limits against their own project requirements.
The more ambitious feature is an AI agent intended to implement back-end functions through chat. In WEVEN’s examples, requests such as adding a bulletin board or reservation system can trigger the creation of the required application functionality and data handling. That could be useful for fast internal tools and smaller customer deployments, but it raises the usual questions around authentication, data protection, database access, auditability, code ownership, and ongoing maintenance.
That makes Zaemit potentially relevant to developers who want an AI web builder integrated with their usual Windows-based VS Code workflow rather than a browser-only no-code environment. It does not eliminate the need for code review or conventional deployment controls, particularly where customer data or production databases are involved.
WEVEN says Zaemit had surpassed 30,000 monthly active users and 300,000 cumulative users by April, and the company plans a broader global push following this professional-service release.
Teams evaluating Zaemit.ai should begin with a non-production project and verify the generated code, hosting model, data controls, and VS Code integration before committing client work.
As reported by IT Dong-A, Zaemit uses specialized agents for planning, design, coding, and back-end work. Users describe what they want in natural language, while the platform assembles a website and its supporting features rather than limiting output to a static page mock-up.
Aimed at production teams, not just prompt-based prototypes
The most substantial additions are a Workspace system, Block Management, and a conversation-driven back-end implementation tool.Workspace is designed as a central dashboard for projects, team members, site status, client communication, domains, and hosting expiration dates. That is a practical feature for agencies and freelancers that routinely juggle renewals and handoffs across several customer sites.
Block Management lets teams save and reuse common interface components or layouts. WEVEN says users can attach detailed AI-generation guidance to blocks, helping preserve a customer’s typography, spacing, colors, and component choices when creating repeat pages or new projects based on an established template.
The company is positioning the service less as an AI site generator and more as a workflow layer that stays involved after the initial page is produced. It also says Zaemit can bring in HTML created with external AI tools for editing and preparation for hosting.
Apollon model and back-end claims
Zaemit.ai introduces an in-house design model called Apollon, described by WEVEN as an evolution of its Maker 3.0/WebGen.Agent technology. The company says the model has been tuned on web-industry classifications and layout systems so it can generate more consistent pages without requiring users to repeatedly specify basic design rules.WEVEN also claims that Apollon reduces token use and generation costs compared with general-purpose large language model workflows. Those are vendor performance claims, and prospective users should test output quality, editability, pricing, and usage limits against their own project requirements.
The more ambitious feature is an AI agent intended to implement back-end functions through chat. In WEVEN’s examples, requests such as adding a bulletin board or reservation system can trigger the creation of the required application functionality and data handling. That could be useful for fast internal tools and smaller customer deployments, but it raises the usual questions around authentication, data protection, database access, auditability, code ownership, and ongoing maintenance.
Windows and VS Code relevance
For Windows developers, the notable point is WEVEN’s existing Visual Studio Code extension. IT Dong-A reports that it supports Model Context Protocol integration, allowing Zaemit to be used from general-purpose AI tools including Claude and ChatGPT. WEVEN plans support for importing externally generated HTML into the Zaemit editor as well.That makes Zaemit potentially relevant to developers who want an AI web builder integrated with their usual Windows-based VS Code workflow rather than a browser-only no-code environment. It does not eliminate the need for code review or conventional deployment controls, particularly where customer data or production databases are involved.
WEVEN says Zaemit had surpassed 30,000 monthly active users and 300,000 cumulative users by April, and the company plans a broader global push following this professional-service release.
Teams evaluating Zaemit.ai should begin with a non-production project and verify the generated code, hosting model, data controls, and VS Code integration before committing client work.
References
- Primary source: 동아일보
Published: 2026-07-20T02:20:34.285005
WEVEN Launches Zaemit.ai for Web Development
- WEVEN introduced 'Zaemit.ai', a B2B multi-agent web production tool. It integrates the entire web creation process into a singl…www.donga.com
