Overview

Today I explored Hugo blog themes extensively while planning a blog refresh, worked through the Vibe Coding Essentials book to sharpen my Claude Code workflow, discovered Archon — a new tool for AI coding workflows — and looked into aiosqlite for async SQLite access in Python.

Highlights

Hugo Theme Deep-Dive — Comparing 10 Themes for a Blog Refresh

I spent time in the Hugo Themes Gallery comparing themes for an upcoming blog refresh. I visited the demo sites for 10 themes and here’s what stood out.

Blog themes:

• PaperMod (★13,116) — The most popular Hugo theme. Fast, clean, and responsive. Supports three layout modes (Regular, Home-Info, Profile), automatic dark/light switching, SEO optimization, and Fuse.js-powered search. No external dependencies like webpack or Node.js required for theme customization — a big plus.
• Stack (★6,261) — A card-style theme built for bloggers. Visually polished layout, with docs in Korean, English, and Chinese. GPL-3.0 license.
• Coder (★3,031) — Simple, clean personal blog theme with dark mode. MIT license.
• Terminal (★2,680) — Retro terminal aesthetic. Great for developers who want personality in their blog.

Documentation and portfolio themes:

• Blox Tailwind (★10,025) — 50+ color themes and widgets included. Works for company sites, portfolios, and blogs.
• Book (★3,953) — Clean book-style documentation theme.
• Docsy (★2,903) — Dedicated theme for technical documentation sites. Apache 2.0 license.
• Compose, Bootstrap — Clean documentation-style and Bootstrap-based themes respectively.

For the deployment side, this Hugo + GitHub Pages guide compares Jekyll, Hexo, and Hugo as static site generators, then covers why Hugo wins on build speed (Go-based, no external dependencies), the GitHub Pages deployment process, setting up Utterances for comments, and managing themes as git submodules.

Archon — Knowledge Hub for AI Coding Assistants

Archon is a knowledge and task management platform for AI coding assistants. It runs as an MCP (Model Context Protocol) server and connects to Claude Code, Cursor, Windsurf, and other AI coding tools.

Core capabilities:

• Knowledge management: Website crawling, PDF/document uploads, automatic code example extraction, vector search-based RAG
• Project/task management: Hierarchical project structure with AI-assisted task creation
• Microservice architecture: Frontend (React+Vite, port 3737), API Server (FastAPI, port 8181), MCP Server (port 8051), Agents (PydanticAI, port 8052)

Spins up with Docker Compose. Uses Supabase (PostgreSQL + PGVector) as the database. Supports OpenAI, Ollama, Google Gemini, and other LLMs, with advanced RAG strategies including hybrid search and result re-ranking.

Cole Medin’s YouTube guide shows real AI coding workflow examples.

Vibe Coding Essentials with Claude Code

Worked through sections 7–10 of Chapter 02 in Weniv Books’ Vibe Coding Essentials. The chapter covers practical development patterns with Claude Code — a hands-on guide to using AI coding tools effectively.

Python aiosqlite — Async SQLite

Studied a guide to aiosqlite for async SQLite access in Python. The standard sqlite3 module is synchronous, which means DB operations block the event loop in an async context. aiosqlite fixes this — it wraps sqlite3 and lets you run DB operations without blocking other coroutines. The API is nearly identical to sqlite3; just add async with and await.

import aiosqlite
import asyncio

async def main():
    async with aiosqlite.connect('example.db') as con:
        cur = await con.cursor()
        await cur.execute('SELECT * FROM stocks WHERE symbol=:symbol', {'symbol': 'RHAT'})
        data = await cur.fetchall()
    print(data)

asyncio.run(main())

Quick Links

• Homebrew + VS Code via Homebrew — macOS dev environment setup
• AWS EC2 (ap-northeast-2) — Frontend deployment and instance management
• AI/dev YouTubers worth following: Cole Medin, Corbin Brown, Rok Benko, Fabio Bergmann

Insights

Two clear threads ran through today’s browsing. First, blog infrastructure renewal — comparing 10 Hugo themes and revisiting the GitHub Pages deployment workflow shows a drive to run the blog more systematically. PaperMod and Stack got the most attention. Second, leveling up the AI coding workflow — exploring Archon, the Vibe Coding Essentials book, and several AI dev YouTubers all point toward moving beyond casual AI tool use toward structured knowledge management and integrated workflows. Archon’s MCP server approach looks particularly useful in environments where multiple AI coding tools are running in parallel.