Software architect focused on enterprise systems and practical ways to work with AI agents.
I'm interested in how people and AI agents can carry a project from design through implementation, documentation, and review—while keeping decisions in human hands and making it clear what changed, why, and what was verified.
I'm currently building Colvia to help that work continue across conversations: preserving project context, approved scope, and reviewable results so each new session can pick up where the last one left off. It builds on my earlier work on ai-workflow-harness.
I write about these experiments, agent collaboration, and lessons from building software on my blog in Korean and English.




