Blog | Tech & Culture Insights

Tech

Client-Server

Traditional Request-Response Model

Tech

WebSocket

Bidirectional Real-time Communication

Tech

SSE (Server-Sent Events)

Server→Client Unidirectional Stream

Tech

GraphQL Subscription

Real-time Events Based on GraphQL Schema

Tech

Webhook

HTTP Callback on Event Trigger

Tech

Publish-Subscribe

Asynchronous Communication via Message Broker

Tech

Message Queue

Asynchronous Task Processing via Queue

Tech

Peer-to-Peer (P2P)

Direct Communication Without a Central Server

Tech

How Does ngrok Work?

Exposing Local Servers to the Internet via Reverse Tunnel

Tech

How Does Slack Socket Mode Work?

Receiving Events via Outbound WebSocket Without Webhooks

Tech

How Does ChatGPT Streaming Work?

Real-time Token-by-Token Response Delivery via SSE

Tech

How Do GitHub Webhooks Work?

git push → HTTP POST → CI/CD Trigger

Tech

How Does Sidekiq Work?

Redis-based Background Job Queue System

Tech

How Does WebRTC Work?

Direct Video/Audio/Data Transfer Between Browsers

Tech

How Does MCP Work?

An Open Protocol Connecting AI Models with External Tools in a Standardized Way

Tech

MCP Transport: stdio vs Streamable HTTP

The Difference Between Local Development and Cloud SaaS Deployment

Tech

How Does tmux Work?

Session Persistence & Splitting with a Terminal Multiplexer

Tech

QUIC vs WebRTC: Why the Switch Back to QUIC?

A Real-world Case of Switching to QUIC After Struggling with WebRTC Complexity

Tech

How Does Location Sensing via WiFi CSI Work?

Using Physical Layer Data of WiFi Signals Like Radar

Tech

What Your Bluetooth Devices Reveal About You

How Life Patterns, Location, and Behavior Are Tracked Through BLE Signals Alone

Ruby on Rails

MVC Pattern

Model-View-Controller — The core architecture of Rails

Ruby on Rails

Convention over Configuration

Rails core philosophy — convention over configuration

Ruby on Rails

Rails Directory Structure

Roles and conventions for each project folder

Ruby on Rails

Rack & Middleware

HTTP processing layer underneath Rails

Ruby on Rails

RESTful Routes

One line of resources generates 7 routes automatically

Ruby on Rails

Params & Strong Parameters

Accessing request parameters and managing allow lists

Ruby on Rails

Before Action & Callbacks

Inserting common logic before/after action execution

Ruby on Rails

Controller Concerns

Modules shared across multiple Controllers

Ruby on Rails

ActiveRecord

ORM — manipulate databases with objects

Ruby on Rails

Migrations

Version control database schema with code

Ruby on Rails

Associations

has_many, belongs_to — defining relationships between models

Ruby on Rails

Validations

Ensure data integrity at the Model level

Ruby on Rails

Model Callbacks

Hooks that auto-execute before/after save/delete

Ruby on Rails

Scopes & Query Interface

Name and manage reusable queries

Ruby on Rails

N+1 Query Problem

Performance enemy — solving with includes

Ruby on Rails

ERB & Partials

Embed Ruby in HTML and split into reusable fragments

Ruby on Rails

Layouts & yield

Manage common page structure with layouts

Ruby on Rails

Form Helpers

Safe and convenient form building with form_with

Ruby on Rails

Turbo & Hotwire

SPA-like UX without JavaScript

Ruby on Rails

Stimulus

HTML-centric lightweight JavaScript framework

Ruby on Rails

Devise

Rails standard authentication system

Ruby on Rails

CSRF Protection

Auto defense against Cross-Site Request Forgery attacks

Ruby on Rails

Strong Parameters

Preventing Mass Assignment attacks

Ruby on Rails

RSpec

BDD (Behavior-Driven Development) test framework

Ruby on Rails

FactoryBot

Clean test data generation

Ruby on Rails

Request Specs

Integration tests for HTTP request/response

Ruby on Rails

Background Jobs

Active Job + Sidekiq — process heavy tasks asynchronously

Ruby on Rails

Action Cable

Rails built-in WebSocket — real-time features

Ruby on Rails

Caching

Dramatically improve response speed by reducing repeated computations

Ruby on Rails

Asset Pipeline & Vite

Frontend asset management for JavaScript, CSS, images

Ruby on Rails

Why We Avoid Helpers

My Convention — data logic in Controller, minimal Helpers

AI Coding Tips

Terminal Setup

Terminal theme and input optimization

AI Coding Tips

Effort Levels

Token usage vs intelligence level control

AI Coding Tips

Plugins, MCP, Skills

MCP server, skills, and agent extensions

AI Coding Tips

Custom Agents

Create custom agents in .claude/agents

AI Coding Tips

Permission Pre-approval

Accelerate workflow with permission pre-approval

AI Coding Tips

Sandbox Mode

Enhanced safety with file/network isolation

AI Coding Tips

Custom Status Line

Display custom info below the composer

AI Coding Tips

Keybinding Customization

Freely remap all keybindings

AI Coding Tips

Hooks

Deterministically intervene in the Claude lifecycle

AI Coding Tips

Spinner Verbs

Customize loading spinner verbs

AI Coding Tips

Output Styles

Configure response tone and format

AI Coding Tips

settings.json Management

37 settings + 84 environment variables

AI Coding Tips

How to Write CLAUDE.md

Pass project-specific context to Claude

AI Coding Tips

Slash Commands & Skills

Quickly execute features with / commands

AI Coding Tips

Auto Memory

Automatic context retention across sessions

AI Coding Tips

/btw (By The Way)

Ask side questions without polluting the main conversation

AI Coding Tips

Auto Permission Mode

Auto-judge tool usage for long autonomous work sessions

AI Coding Tips

/voice — Voice Input

Give instructions to Claude Code by voice

AI Coding Tips

.cursorrules File

Pass project rules to Cursor

AI Coding Tips

Composer (Agent Mode)

Multi-file automatic editing

AI Coding Tips

Tab Completion

Optimized code auto-completion

AI Coding Tips

.github/copilot-instructions.md

Pass project rules to Copilot

AI Coding Tips

Chat Agents (@workspace)

Use the entire codebase as context

AI Coding Tips

Copilot Workspace

Issue → Plan → Code automatic workflow

AI Coding Tips

Project Context File Comparison

CLAUDE.md vs .cursorrules vs copilot-instructions

AI Coding Tips

Effective Prompt Writing

Give clear instructions to AI coding agents

AI Coding Tips

Context Window Management

Work efficiently within token limits

AI Coding Tips

Automate Test Writing with AI

Request tests right after writing code

AI

Neural Network

Connectionist learning model inspired by neurons

AI

Transformer

Parallel processing architecture based on Self-Attention

AI

Pre-training

The stage of learning general knowledge from massive data

AI

Fine-tuning

Re-training a pre-trained model for specific tasks

AI

RLHF

Reinforcement Learning from Human Feedback

AI

DPO (Direct Preference Optimization)

Direct preference optimization without a reward model

AI

LoRA (Low-Rank Adaptation)

Efficient fine-tuning that trains only a small number of parameters

AI

Prompt Engineering

Maximizing LLM capabilities through prompt design

AI

RAG (Retrieval-Augmented Generation)

Improving accuracy by injecting external knowledge through retrieval

AI

Chain-of-Thought (CoT)

Solving complex problems through step-by-step reasoning

AI

Function Calling (Tool Use)

LLMs call external tools to perform real-world tasks

AI

Quantization

Lightening models by converting weights to lower precision

AI

Knowledge Distillation

Transferring knowledge from large models to small models

AI

Mixture of Experts (MoE)

Efficient scaling by activating only needed experts

AI

Speculative Decoding

Small model drafts predictions, large model verifies

AI

Vision-Language Model

AI that understands both images and text together

AI

Diffusion Model

Generating images by progressively removing noise

AI

Text-to-Speech (TTS)

Converting text to natural-sounding speech

AI

AI Agent

LLMs autonomously plan and use tools to complete tasks

AI

MCP (Model Context Protocol)

Standard protocol connecting AI to external tools

AI

Eval-Driven Development

Iteratively improving AI systems based on evaluation (Eval)

AI

AI Alignment

Making AI act in accordance with human intent and values

AI

Constitutional AI (CAI)

A method to improve AI itself using a constitution (principles)

AI

Hallucination

The phenomenon of AI generating plausible but false content

AI

Audio Tokenization (Neural Audio Codec)

Converting audio into discrete tokens for LLM-like processing

AI

Evolution of Speech Models

Rule-based → Deep learning pipeline → Token-based → Unified multimodal

AI

GPT-SoVITS Practical Analysis

How Pre-training and Fine-tuning work in actual code