Our Story

As a child, I dreamed of becoming an astronaut. What stayed with me was not only the idea of spaceflight, but the harder question behind it: how do we protect human life in an uncertain and unforgiving environment? Studying spacecraft environment and life-support engineering became my first encounter with the systems that quietly stand behind every powerful technology.

My career continued across aerospace systems, rail transit signaling, automotive functional safety, semiconductor functional safety, and international safety certification. The industries changed, but the engineering discipline remained constant: identify hazards, establish boundaries, design intervention, and build verifiable evidence that a system can be trusted.

AI is moving from answering questions to taking action. Agents call tools and access operational systems. Robots enter factories, hospitals, homes, and transport infrastructure. The central question is no longer only whether AI is capable, but whether its actions are authorized, bounded, monitored, recoverable, and traceable.

This is why HMsafe exists. We build the safety control layer between AI decisions, human oversight, and physical-world execution. Through authorization, execution boundaries, monitoring, human override, simulation validation, minimum risk states, and audit evidence, we help organizations release AI capability while keeping potential harm understandable, controllable, and traceable.

The childhood dream never disappeared; the field of responsibility changed. I once wanted to protect people traveling into space. Today, I want to protect everyone who will live and work alongside intelligent systems. HMsafe carries the discipline of life support and system safety into the intelligent era.