Frontend Engineer who builds tools to eliminate inefficiency.
- π’ Frontend Engineer @ Naver β Clip Web
- π Building Engram β AI-powered memory engine for Obsidian
- π± Exploring Agentic Workflows, MCP, AI-native Development
- βοΈ Writing at jegwan.hashnode.dev
| Project | Description |
|---|---|
| Agent Anatomy | Interactive visualization of LLM agent internals β real API message flow. Agent Engineering = Context Engineering. |
| Engram | MCP server for Obsidian vault augmentation β FTS5, semantic search, knowledge graph |
| Cosmos | Custom AI agent framework WIP |
| Intent Sharpening | Claude Code plugin β Socratic questioning to sharpen vague intentions into clear Intent Cards (6-axis clarity scoring) |
| Team Context | Shared team knowledge mounting system for AI agents β mount/unmount/sync team docs into Claude Code |
| gongmyung-kanban | Lifecycle-tracking kanban with aging badges & WIP limits |
20+ custom skills & hooks I built for my AI-augmented workflow:
| Category | Examples |
|---|---|
| Daily Ops | /briefing /organize /νμΈ β morning briefing, inbox triage, situation awareness |
| Knowledge | /study /feed /tech-news β Socratic learning sessions, content digestion, tech news curation |
| People | /dossier /1on1 β deep profile research, structured 1on1 notes |
| Publishing | /wiki /blog /upload-ai-board β one-command publishing to Confluence, Hashnode, internal boards |
| Work | /work /task /jira-create β kanban auto-execution, task orchestration, Jira integration |
| Infra | /schedule cron scheduler, session-end hooks, ABC loop tracking |
