▲ At the Intellicon Solutions EgentWrX launch event, from left: KENMEC Group Chairman Kevin Hsieh (謝明凱), AWS Senior Go-to-Market Manager for Generative AI Matt Yang (楊書維), Intellicon Solutions Chief Operating Officer Claudia Chen (陳旻筠), Intellicon Solutions Chief Executive Officer Albert Chen (陳慶蔚), and New Wide Group Executive Director Sunny Huang (黃冠華).
Intellicon Solutions officially launched EgentWrX, an enterprise-grade Agentic AI platform, on September 22. Designed to be easy for employees to use, the platform also provides comprehensive enterprise management mechanisms. EgentWrX gives every employee a dedicated AI Agent equipped with enterprise knowledge, memory, skills, and permissions, while Agentic Workflow connects tasks that were previously dispersed across different roles and departments. The platform supports SaaS cloud, on-premises, and hybrid private-cloud deployments to serve the growing AI transformation needs of manufacturers in industries including textiles, machining, metal fasteners, and automotive components.
AI Agents Have Become Workplace Partners, Bringing Management Challenges: How Many AI Agents Are Actually Operating Within the Company?
Over the past two years, AI Agents have entered enterprises considerably faster than management frameworks have been established. According to survey findings, 61.8% of AI applications within enterprises operate entirely outside corporate oversight. These applications, commonly known as “shadow AI,” pose management and security risks to enterprises. Intellicon Solutions Chief Executive Officer Albert Chen emphasized: “AI Agents have indeed made some work faster, but from a manager’s perspective, most enterprises cannot readily answer the question of how effective their AI implementation has actually been. AI is developing so quickly that everyone is still exploring how to apply it, let alone how to manage it. What enterprises lack today is therefore not a more powerful model, but an operating framework that brings AI Agents under organizational management.”
To address this issue, EgentWrX takes a management-oriented approach by providing an AI-native operating platform aligned with an enterprise’s existing organizational structure: each position is assigned an AI Agent, each repetitive piece of work becomes a Task, and each working method is organized into a reusable Skill. The three layers—Agent, Task, and Skill—correspond respectively to positions, duties, and competencies, while AI Agents follow the enterprise’s own knowledge, documents, and management methods. Knowledge assets shared across the company or a department can be converted into Skills that everyone in the company can access according to their permissions. Working methods accumulated over many years no longer remain solely in the experience of veteran employees.
Understanding Organizational Structures and Corporate Rules: Every AI Agent Has an Identity, Permissions, and a Budget
In addition to aligning with an enterprise’s existing operating structure, EgentWrX provides comprehensive access controls that can be configured by department, role, and data classification, and can integrate with the enterprise’s existing AD and SSO systems. For Token management, enterprises can manage Token usage at multiple levels, including by department and individual, enabling early warnings and timely responses. The activity history of every AI Agent is also traceable and available for analysis. Intellicon Solutions Chief Operating Officer Claudia Chen further explained: “Managers can clearly see which work matters AI Agents have begun handling and review their execution counts, frequency, efficiency, and outcomes. When evaluating the effectiveness of AI in the future, managers will be able to do more than say, ‘My company has implemented AI.’ They will be able to see at a glance exactly which work AI is being used for and determine whether the resulting Token consumption is genuinely reducing costs and increasing efficiency.”
▲ Intellicon Solutions unveiled EgentWrX, its new enterprise-grade Agentic AI management platform. The platform is designed to give every employee a dedicated AI Agent and connect these AI Agents to build an enterprise AI team. Nearly 50 companies and more than 100 enterprise partners attended the event.
First Help AI Agents Understand the Company, Then Turn Company Knowledge into Real Productivity
General-purpose models do not know an enterprise’s internal quotation procedures, quality inspection standards, or special customer terms. Enabling AI Agents to acquire the knowledge and experience of senior employees therefore remains essential. EgentWrX provides multiple complementary knowledge bases, including vector, LLM Wiki, and SQL options, to address different needs. Vector-based semantic retrieval processes large volumes of information and is suitable for technical documents, operating procedures, and similar materials. LLM Wiki enables AI Agents to understand “the complete context surrounding people, events, and things.” EgentWrX also includes a built-in data-cleaning module that helps process unstructured data that is unsuitable for direct use by AI Agents. Enterprises can organize years of accumulated working methods into knowledge assets that AI Agents can understand and retrieve without having to assemble their own data engineering teams.
At the same time, enterprises depend on various SQL databases for daily operations: preparing quotations requires checking historical prices, accepting orders requires reviewing inventory, and shipping requires confirming production schedules. EgentWrX uses MCP to integrate an enterprise’s internal databases and applications, allowing AI Agents to directly query the latest data within the scope of their permissions and creating an AI Agent that “understands both enterprise knowledge and real-world data.”
AI Agents Hand Work Off to One Another, Reducing Inquiry Processing Time from Eight Hours to 28 Minutes
A single AI Agent can help one person work faster, but it is difficult for that Agent to improve the efficiency of cross-departmental work. Lengthy business processes invariably contain several “links that have not yet been AI-enabled.” EgentWrX allows AI Agents to hand off work results to one another, forming an Agentic Workflow. In one inquiry process, for example, after a sales AI Agent receives an inquiry email, it can automatically send requests to AI Agents in the procurement, finance, and production management departments. Each AI Agent then uses its department’s knowledge and permissions to perform cost estimates and reviews, and confirm delivery dates. The results are returned to the sales AI Agent and compiled into a quotation. In this case, the total processing time was reduced from eight hours to 28 minutes.
Intellicon Solutions’ FDE Team Supports Enterprises in Deploying AI Agent Teams in Phases
Intellicon Solutions has an FDE (Forward Deployed Engineer)team with cumulative experience supporting AI implementation at more than 100 enterprises. The team will support enterprises through three stages to establish their own capabilities: inventorying their organizations and work, enabling employees to create Tasks themselves, and organizing working methods into company Skills, thereby building sustainable, self-sufficient enterprise AI capabilities.
Albert Chen said: “During the first wave of digital transformation, enterprises moved their processes into ERP systems. In the AI era, what needs to be digitized is how people work. EgentWrX gives every position an AI Agent that understands the company’s rules and knows how to complete each task. Most importantly, it can be managed and audited. For an enterprise, AI has truly evolved from a personal tool into organizational productivity only when its leaders can confidently state how many AI Agents are currently operating within the company and what each one has helped accomplish.”