Preserve engineering know-how at scale

Industrial manufacturing teams depend on senior engineers to diagnose issues, validate fixes, and keep production quality stable. Their expertise is built from years of understanding how equipment behaves and recovers from failures. When those experts are unavailable, newer engineers take longer to troubleshoot and validate the right fixes.

As Head of Product & Design, I led product and design direction of aiKO Knowledge Base, turning senior engineering expertise into AI-usable knowledge.

I modeled the product after how engineers organize and recall diagnostic knowledge: breaking equipment into components, linking symptoms to failure patterns, and connecting likely causes to proven resolutions. This became the agent's memory layer.

I also designed the interaction model between humans, agents, and industrial systems. Senior engineers captured knowledge. Machines and APIs supplied real-time and historical context. The agent used those inputs to trace a symptom to a cause and a solution backed by verified knowledge. Junior engineers could apply that guidance during incidents and submit new evidence for expert review.

PepsiCo, RHI Magnesita, and Tokyo Electron Limited adopted the solution as early design partners, testing the model across different equipment and operating environments. Workflow modeling with customer teams projected 25% faster troubleshooting and 40% faster time to independent troubleshooting for new engineers.