Notes from building backend, AI, and product systems.https://ynaito.dev/en-UShttps://ynaito.dev/en/writing/vibe-coding-for-beginners/https://ynaito.dev/en/writing/vibe-coding-for-beginners/Learn what vibe coding originally meant, how it differs from responsible AI-assisted development, and how to build a study timer by talking to an AI. This beginner-friendly guide covers tool selection, reusable prompts, hands-on testing, and the risks of handling personal data, credentials, and production systems.Sat, 11 Jul 2026 16:33:50 GMTAI AgentGenerative AIhttps://ynaito.dev/en/writing/codex-gpt-5-6-sol-terra-luna/https://ynaito.dev/en/writing/codex-gpt-5-6-sol-terra-luna/Learn how to update Codex CLI for GPT-5.6, switch between Sol, Terra, and Luna, and configure reasoning effort independently. This guide also presents a reproducible comparison protocol covering quality, latency, usage, and human rework, followed by practical model-selection guidance for common engineering tasks.Sat, 11 Jul 2026 16:08:04 GMTAI AgentCodexhttps://ynaito.dev/en/writing/engineering-in-the-ai-agent-era/https://ynaito.dev/en/writing/engineering-in-the-ai-agent-era/AI agents dramatically reduce implementation costs, but they do not remove responsibility for delivering correct software safely. This article explains why engineering value is shifting from code production toward constraint design, context engineering, assumption review, and evidence-based verification.Sat, 11 Jul 2026 15:52:35 GMTAI Agenthttps://ynaito.dev/en/writing/codex-agent-loop-boundary/https://ynaito.dev/en/writing/codex-agent-loop-boundary/This article decomposes the mechanism that lets Codex CLI inspect, edit, test, and revise software into four responsibilities: model, harness, tools, and environment. It follows the tool-call boundary through execution feedback, permissions, MCP, termination, and context management.Sat, 11 Jul 2026 15:18:19 GMTAI AgentGenerative AIhttps://ynaito.dev/en/writing/inside-ai-coding-agents/https://ynaito.dev/en/writing/inside-ai-coding-agents/An AI coding agent does not generate a complete solution in one shot. It combines a model, context management, tools, an execution environment, and safety controls in a repeated cycle of deciding, acting, and observing. This article follows a practical API change to explain that architecture, its failure modes, and how to use coding agents safely and effectively.Sat, 11 Jul 2026 14:18:09 GMTAI Agenthttps://ynaito.dev/en/writing/codex-cli-toha-chat-ai-chigai/https://ynaito.dev/en/writing/codex-cli-toha-chat-ai-chigai/Codex CLI does more than answer coding questions. It can inspect project files, edit code, run commands, and verify the results. This beginner-friendly guide explains how a coding agent differs from conventional chat-based AI.Sat, 11 Jul 2026 07:55:56 GMTGenerative AI