Computer Science & Engineering · Seoul, Korea

Jaeseok An

I build tools that turn ambiguous work into workflows people and AI agents can actually execute.

GitHub ↗ LinkedIn ↗ Email ↗
Selected Work

e-Class MCP

GitHub ↗
Open Source MCP Claude Code

Turning Chung-Ang University's browser-bound LMS into an agent-accessible tool layer.

  • Built an open-source MCP server for university workflows.
  • Exposes course materials, assignments, grades, and syllabus data to AI agents.
  • Dogfooded across six current courses.

Education Content System

Visit ↗
AI Workflow SEO Cloudflare

A human-in-the-loop publishing system for Korean mathematics solutions, built around AI-assisted production and web distribution.

LLM draft → math formatting → SEO-ready HTML → human review
91K+ cumulative page views
5.1K peak daily page views
~5× Apr–Aug growth

AeightPlus

Product Leadership Team Design

Designed a team-formation service around a simple observation: strong individuals can still underperform when their working styles do not fit.

  • Led a five-person, cross-year university team.
  • Developed the project from concept through Figma and final presentation over 16 weeks.
  • Presented the result to approximately 100 students.
Community

GDG on Campus, Chung-Ang University

Selected member of the university developer community.

2026–27

SearchOS Codex Skillathon

Invited panelist for a session with approximately 50–60 professionals.

PANELIST

Peer Claude Code Mentoring

Helped students move from chat-based AI use toward practical coding-agent workflows.

PEER-LED

I’m especially interested in what happens when AI stops being a place you visit and becomes part of the tools and workflows you already use.