Coming from YouTube?
Welcome,
I'm Dante.
The AI coding workflows and frontend practice from the videos, taken all the way here by building them yourself. Start with a free course, and work through the hard parts together in 1:1 mentoring.
Three ways in
Pick the entry point that fits you now
Learn with a course
밑바닥부터 만드는 AI 에이전트 — RAG · 멀티 에이전트 · MCP
프레임워크 없이 Python과 while 문 하나로 AI 에이전트를 직접 만듭니다. 입문에서는 도구·실패와 비용·기억·사람 승인·평가·웹 화면을 한 편씩 붙여 코딩 에이전트를 완성하고, 실전에서는 실제 상장사 사업보고서로 RAG · 멀티 에이전트 · MCP 서버를 구현해 AI 투자 리서치 워크벤치를 완성합니다. 마지막 AI 평가 파트에서는 그 에이전트를 평가 세트 · 루브릭 · LLM 판사 · 통계 · CI 회귀 관문으로 직접 평가하며 AI 평가 · AI 엔지니어 면접까지 준비합니다.
Enroll₩80,000
1:1 mentoring
1:1 mentoring on your own code
Dante, a working developer, reviews your code and resume himself. Choose a single 60-minute session or the 8-session intensive.
From ₩150,000Learn more
Dante beyond the videos
Taught the way it is done at work
The judgment calls the videos only sketch are shown in full in courses and mentoring, with real code and reviews.
- Work
- Engineer at a top Korean tech company
- Senior experience, passed on as is
- Teaching
- Hands-on code review
- Your assignments and code, reviewed directly
Dante on YouTube @danteDeveloper
Latest videos
Meet the AI coding and frontend practice on video first. New uploads show up here automatically.
Dante Learn Blog
Blog
Write-ups of what we build and measure in our courses.
PHP's Emergency Patch: Can Servers Be Taken Over Without a Login? How PHP2Shell Works and How to Check and Patch Your Server
On October 21, PHP shipped an emergency update fixing two critical vulnerabilities that allow unauthenticated remote code execution. We explain the two holes the security industry calls PHP2Shell with simple analogies, then walk through checking the version, applying a temporary mitigation, and upgrading to 8.4.27 on a local Docker server.
Claude Code and Codex on One Screen: Splitting Work Between Two Agents and Having Them Review Each Other with Herdr
Once you run two or three agents, you start losing track of who is working and who is waiting on you. We used Herdr, a terminal workspace for agents, to run Claude Code and Codex side by side, split work across git worktrees, and get a review from a different model, building a real dashboard of public GitHub issues along the way.
Every Test Passed, and There Was Still a Bug: Finding Ordering Bugs with pstack and TLA+
We installed pstack, the agent skill set a Cursor engineer uses every day, into Claude Code and used formal verification (TLA+) to find and fix a bug the tests missed. Then we opened the open-source code line by line and matched "this part works like this, so the agent did that" against the real session logs.
Tracks
Choose by what you walk away with
Career build-up
Junior to mid-level frontend and web developers
- A portfolio you can explain in interviews
- Code review history from a working engineer
- Practical React and frontend patterns
Shipping products and landing pages
Solo developers, side projects, second jobs
- A live service that takes real payments
- Landing, checkout, onboarding, and retention flows
- Analytics that show your conversion
- Open now
AI development workflow
Developers at every level
- Using Codex, specs, and agents
- Your own AI pair programming setup
- AI tools you understand, not just use
1:1 mentoring
The answers a comment can't give,
one on one.
"Why doesn't it work in my code?" That question needs someone to look at the code. We go through your assignments and code over online 1:1 calls and decide what to fix next every session.
See mentoring- Format
- Online 1:1 video calls
- Options
- Single session · 8-session intensive
- Feedback
- Assignments and code review
- Price
- From ₩150,000