6 Months of Vibe Coding: I Built a Wireframe MCP
“No matter how well I explained in text, it couldn’t replace a screen.” The e-torch project was progressing smoothly. But as I explained screens in more det...
Tags
- AI 14
- Claude Code 14
- Automation 10
- Prompt Engineering 9
- Agent 9
- Productivity 2
- vue-pivottable 1
- Vue 1
- Learning 1
- Side Project 1
- Vibe Coding 1
- Developer Experience 1
- Dev Tools 1
- React 1
- tailwind-grid-layout 1
- Open Source 1
- DSL 1
- Wireweave 1
- Wireframe 1
- MCP 1
AI
6 Months of Vibe Coding: 6-Year-Old Issues Resolved in a Week
“I’ll do it someday.” There were issues I had been putting off for 6 years. I resolved all 12 of them in a week.
6 Months of Vibe Coding: Automating One Annoyance at a Time
“When I find something inconvenient, automating it has become a pattern.” I wrote this in my 6-month retrospective. Without any grand plan, I just built too...
6 Months of Vibe Coding: I Thought I Was Getting Dumber
“Am I getting dumber?” A few months after I started using Claude Code, this thought suddenly hit me. AI writes the code, and I just review it. The feeling o...
[AI Agent Pipeline #10] Completed App and Retrospective
In the previous article, we covered improvements discovered while operating the pipeline. This is the final article. Started in August 2025, six months late...
[AI Agent Pipeline #9] Pipeline Simplification
In the previous article, we covered the retry and rollback mechanism. This article summarizes improvements discovered while operating the pipeline.
[AI Agent Pipeline #8] Retry and Rollback
In the previous article, we covered the normal execution flow of the pipeline. This article summarizes how I handled failures.
[AI Agent Pipeline #7] Running Agents with Shell Scripts
In the previous article, we covered the collaboration structure of 7 agents and the handoff protocol. This article summarizes the shell script orchestration...
[AI Agent Pipeline #6] How 7 Agents Collaborate
In the previous article, we covered the process of completing agent prompts. While discussing prompts, I talked about “failure” and “success”—those results w...
[AI Agent Pipeline #5] 4 Stages to Completing the Prompt
In the previous article, we covered the story of deciding to apply the AI-DLC methodology when stuck. This article summarizes how I completed the prompts wh...
[AI Agent Pipeline #4] Why It Still Didn’t Work After Splitting
In the previous article, we covered why a single prompt didn’t work and the process of splitting into 7 agents. This article covers what I tried after split...
[AI Agent Pipeline #3] Why a Single Prompt Didn’t Work
In the previous article, we used the metadata pipeline to automatically generate category.yaml and empty content files. Starting from this article, we’ll co...
[AI Agent Pipeline #2] What to Generate
In the previous post, I talked about why I started building a learning app. This post covers how I defined the content to generate.
[AI Agent Pipeline #1] Why I Started Building a Learning App
Have you ever gotten tired of studying with LLMs? I did. So I decided to auto-generate learning content with LLM and turn it into an app. This series do...
Claude Code
6 Months of Vibe Coding: I Built a Wireframe MCP
“No matter how well I explained in text, it couldn’t replace a screen.” The e-torch project was progressing smoothly. But as I explained screens in more det...
6 Months of Vibe Coding: 6-Year-Old Issues Resolved in a Week
“I’ll do it someday.” There were issues I had been putting off for 6 years. I resolved all 12 of them in a week.
6 Months of Vibe Coding: Automating One Annoyance at a Time
“When I find something inconvenient, automating it has become a pattern.” I wrote this in my 6-month retrospective. Without any grand plan, I just built too...
6 Months of Vibe Coding: I Thought I Was Getting Dumber
“Am I getting dumber?” A few months after I started using Claude Code, this thought suddenly hit me. AI writes the code, and I just review it. The feeling o...
[AI Agent Pipeline #10] Completed App and Retrospective
In the previous article, we covered improvements discovered while operating the pipeline. This is the final article. Started in August 2025, six months late...
[AI Agent Pipeline #9] Pipeline Simplification
In the previous article, we covered the retry and rollback mechanism. This article summarizes improvements discovered while operating the pipeline.
[AI Agent Pipeline #8] Retry and Rollback
In the previous article, we covered the normal execution flow of the pipeline. This article summarizes how I handled failures.
[AI Agent Pipeline #7] Running Agents with Shell Scripts
In the previous article, we covered the collaboration structure of 7 agents and the handoff protocol. This article summarizes the shell script orchestration...
[AI Agent Pipeline #6] How 7 Agents Collaborate
In the previous article, we covered the process of completing agent prompts. While discussing prompts, I talked about “failure” and “success”—those results w...
[AI Agent Pipeline #5] 4 Stages to Completing the Prompt
In the previous article, we covered the story of deciding to apply the AI-DLC methodology when stuck. This article summarizes how I completed the prompts wh...
[AI Agent Pipeline #4] Why It Still Didn’t Work After Splitting
In the previous article, we covered why a single prompt didn’t work and the process of splitting into 7 agents. This article covers what I tried after split...
[AI Agent Pipeline #3] Why a Single Prompt Didn’t Work
In the previous article, we used the metadata pipeline to automatically generate category.yaml and empty content files. Starting from this article, we’ll co...
[AI Agent Pipeline #2] What to Generate
In the previous post, I talked about why I started building a learning app. This post covers how I defined the content to generate.
[AI Agent Pipeline #1] Why I Started Building a Learning App
Have you ever gotten tired of studying with LLMs? I did. So I decided to auto-generate learning content with LLM and turn it into an app. This series do...
Automation
6 Months of Vibe Coding: Automating One Annoyance at a Time
“When I find something inconvenient, automating it has become a pattern.” I wrote this in my 6-month retrospective. Without any grand plan, I just built too...
[AI Agent Pipeline #10] Completed App and Retrospective
In the previous article, we covered improvements discovered while operating the pipeline. This is the final article. Started in August 2025, six months late...
[AI Agent Pipeline #9] Pipeline Simplification
In the previous article, we covered the retry and rollback mechanism. This article summarizes improvements discovered while operating the pipeline.
[AI Agent Pipeline #8] Retry and Rollback
In the previous article, we covered the normal execution flow of the pipeline. This article summarizes how I handled failures.
[AI Agent Pipeline #7] Running Agents with Shell Scripts
In the previous article, we covered the collaboration structure of 7 agents and the handoff protocol. This article summarizes the shell script orchestration...
[AI Agent Pipeline #6] How 7 Agents Collaborate
In the previous article, we covered the process of completing agent prompts. While discussing prompts, I talked about “failure” and “success”—those results w...
[AI Agent Pipeline #5] 4 Stages to Completing the Prompt
In the previous article, we covered the story of deciding to apply the AI-DLC methodology when stuck. This article summarizes how I completed the prompts wh...
[AI Agent Pipeline #4] Why It Still Didn’t Work After Splitting
In the previous article, we covered why a single prompt didn’t work and the process of splitting into 7 agents. This article covers what I tried after split...
[AI Agent Pipeline #3] Why a Single Prompt Didn’t Work
In the previous article, we used the metadata pipeline to automatically generate category.yaml and empty content files. Starting from this article, we’ll co...
[AI Agent Pipeline #2] What to Generate
In the previous post, I talked about why I started building a learning app. This post covers how I defined the content to generate.
Prompt Engineering
[AI Agent Pipeline #10] Completed App and Retrospective
In the previous article, we covered improvements discovered while operating the pipeline. This is the final article. Started in August 2025, six months late...
[AI Agent Pipeline #9] Pipeline Simplification
In the previous article, we covered the retry and rollback mechanism. This article summarizes improvements discovered while operating the pipeline.
[AI Agent Pipeline #8] Retry and Rollback
In the previous article, we covered the normal execution flow of the pipeline. This article summarizes how I handled failures.
[AI Agent Pipeline #7] Running Agents with Shell Scripts
In the previous article, we covered the collaboration structure of 7 agents and the handoff protocol. This article summarizes the shell script orchestration...
[AI Agent Pipeline #6] How 7 Agents Collaborate
In the previous article, we covered the process of completing agent prompts. While discussing prompts, I talked about “failure” and “success”—those results w...
[AI Agent Pipeline #5] 4 Stages to Completing the Prompt
In the previous article, we covered the story of deciding to apply the AI-DLC methodology when stuck. This article summarizes how I completed the prompts wh...
[AI Agent Pipeline #4] Why It Still Didn’t Work After Splitting
In the previous article, we covered why a single prompt didn’t work and the process of splitting into 7 agents. This article covers what I tried after split...
[AI Agent Pipeline #3] Why a Single Prompt Didn’t Work
In the previous article, we used the metadata pipeline to automatically generate category.yaml and empty content files. Starting from this article, we’ll co...
[AI Agent Pipeline #2] What to Generate
In the previous post, I talked about why I started building a learning app. This post covers how I defined the content to generate.
Agent
[AI Agent Pipeline #10] Completed App and Retrospective
In the previous article, we covered improvements discovered while operating the pipeline. This is the final article. Started in August 2025, six months late...
[AI Agent Pipeline #9] Pipeline Simplification
In the previous article, we covered the retry and rollback mechanism. This article summarizes improvements discovered while operating the pipeline.
[AI Agent Pipeline #8] Retry and Rollback
In the previous article, we covered the normal execution flow of the pipeline. This article summarizes how I handled failures.
[AI Agent Pipeline #7] Running Agents with Shell Scripts
In the previous article, we covered the collaboration structure of 7 agents and the handoff protocol. This article summarizes the shell script orchestration...
[AI Agent Pipeline #6] How 7 Agents Collaborate
In the previous article, we covered the process of completing agent prompts. While discussing prompts, I talked about “failure” and “success”—those results w...
[AI Agent Pipeline #5] 4 Stages to Completing the Prompt
In the previous article, we covered the story of deciding to apply the AI-DLC methodology when stuck. This article summarizes how I completed the prompts wh...
[AI Agent Pipeline #4] Why It Still Didn’t Work After Splitting
In the previous article, we covered why a single prompt didn’t work and the process of splitting into 7 agents. This article covers what I tried after split...
[AI Agent Pipeline #3] Why a Single Prompt Didn’t Work
In the previous article, we used the metadata pipeline to automatically generate category.yaml and empty content files. Starting from this article, we’ll co...
[AI Agent Pipeline #2] What to Generate
In the previous post, I talked about why I started building a learning app. This post covers how I defined the content to generate.
Productivity
6 Months of Vibe Coding: Automating One Annoyance at a Time
“When I find something inconvenient, automating it has become a pattern.” I wrote this in my 6-month retrospective. Without any grand plan, I just built too...
6 Months of Vibe Coding: I Thought I Was Getting Dumber
“Am I getting dumber?” A few months after I started using Claude Code, this thought suddenly hit me. AI writes the code, and I just review it. The feeling o...
vue-pivottable
6 Months of Vibe Coding: 6-Year-Old Issues Resolved in a Week
“I’ll do it someday.” There were issues I had been putting off for 6 years. I resolved all 12 of them in a week.
Vue
6 Months of Vibe Coding: 6-Year-Old Issues Resolved in a Week
“I’ll do it someday.” There were issues I had been putting off for 6 years. I resolved all 12 of them in a week.
Learning
[AI Agent Pipeline #1] Why I Started Building a Learning App
Have you ever gotten tired of studying with LLMs? I did. So I decided to auto-generate learning content with LLM and turn it into an app. This series do...
Side Project
[AI Agent Pipeline #1] Why I Started Building a Learning App
Have you ever gotten tired of studying with LLMs? I did. So I decided to auto-generate learning content with LLM and turn it into an app. This series do...
Vibe Coding
6 Months of Vibe Coding: I Thought I Was Getting Dumber
“Am I getting dumber?” A few months after I started using Claude Code, this thought suddenly hit me. AI writes the code, and I just review it. The feeling o...
Developer Experience
6 Months of Vibe Coding: I Thought I Was Getting Dumber
“Am I getting dumber?” A few months after I started using Claude Code, this thought suddenly hit me. AI writes the code, and I just review it. The feeling o...
Dev Tools
6 Months of Vibe Coding: Automating One Annoyance at a Time
“When I find something inconvenient, automating it has become a pattern.” I wrote this in my 6-month retrospective. Without any grand plan, I just built too...
React
6 Months of Vibe Coding: 6-Year-Old Issues Resolved in a Week
“I’ll do it someday.” There were issues I had been putting off for 6 years. I resolved all 12 of them in a week.
tailwind-grid-layout
6 Months of Vibe Coding: 6-Year-Old Issues Resolved in a Week
“I’ll do it someday.” There were issues I had been putting off for 6 years. I resolved all 12 of them in a week.
Open Source
6 Months of Vibe Coding: 6-Year-Old Issues Resolved in a Week
“I’ll do it someday.” There were issues I had been putting off for 6 years. I resolved all 12 of them in a week.
DSL
6 Months of Vibe Coding: I Built a Wireframe MCP
“No matter how well I explained in text, it couldn’t replace a screen.” The e-torch project was progressing smoothly. But as I explained screens in more det...
Wireweave
6 Months of Vibe Coding: I Built a Wireframe MCP
“No matter how well I explained in text, it couldn’t replace a screen.” The e-torch project was progressing smoothly. But as I explained screens in more det...
Wireframe
6 Months of Vibe Coding: I Built a Wireframe MCP
“No matter how well I explained in text, it couldn’t replace a screen.” The e-torch project was progressing smoothly. But as I explained screens in more det...
MCP
6 Months of Vibe Coding: I Built a Wireframe MCP
“No matter how well I explained in text, it couldn’t replace a screen.” The e-torch project was progressing smoothly. But as I explained screens in more det...