Knowledge Base in aiKO

Storage that mimics how humans think

Company

Aitomatic

Position

Product and Design Lead

Year

2024

Market

Asia and the U.S., B2B

Challenges

In manufacturing, senior engineers hold the keys to keeping operations running smoothly. They know how machines behave, how to troubleshoot failures, and what processes work best. But this expertise rarely lives in one place. Scattered across spreadsheets, emails, and informal chats, it becomes fragmented and, worse, disappears when these experts leave.

Production slows, problems take longer to solve, and companies face financial losses.

Solution

As Product and Design Lead at Aitomatic, I led the creation of the Knowledge Base in aiKO—a function that preserves expertise and turns it into real-time, actionable intelligence.

Traditional systems force people to work against how they naturally think. I hypothesized a new approach: align knowledge organization with how people process and connect ideas. I directed the product team—two front-end engineers, two back-end engineers, and a junior designer—to test and apply this theory. We studied how people link concepts and mapped these insights to the complex workflows of industrial engineers.

The breakthrough was a system that mirrors how engineers operate. Information is structured hierarchically, like a mind map, with clear relationships between guides, workflows, and best practices. It’s a living, connected network that makes knowledge actionable. But the real power came from integrating structured knowledge with live machine data. A troubleshooting guide for a motor failure, for example, doesn’t just outline steps—it pulls real-time sensor readings to show exactly what’s happening. Engineers can act faster, with clarity and precision, solving problems before they escalate.

I worked closely with the team to balance speed and quality: simplifying the user experience with the front-end engineers, ensuring seamless integration with the back-end engineers, and mentoring the junior designer to deliver a clear, functional UI that streamlined complex workflows.

Knowledge Hierarchy

Knowledge Hierarchy

Knowledge Hierarchy

Entities

Entities

Entities

Impact

While still in its early stages, the Knowledge Base is projected to reduce troubleshooting time by 25% and accelerate onboarding for new engineers by 40%.

Our customers like Pepsi, RHI Magnesita, and Tokyo Electron recognized its potential to stabilize operations, improve collaboration and set a foundation for smarter operations where teams solve problems faster and with confidence.

Knowledge Domain

Knowledge Domain

Knowledge Domain

Add Knowledge

Add Knowledge

Add Knowledge

Modify Expert's Plan

Modify Expert's Plan

Modify Expert's Plan

File Uploaded

File Uploaded

File Uploaded

Agents Linked

Agents Linked

Agents Linked

File Citations in Chat

File Citations in Chat

File Citations in Chat

Save Memory

Save Memory

Save Memory

Memory in Chat

Memory in Chat

Memory in Chat

Integrating Data

Integrating Data

Integrating Data

Connect to Internet

Connect to Internet

Connect to Internet

Connect to API

Connect to API

Connect to API

Data Citations in Chat

Data Citations in Chat

Data Citations in Chat