Tag
一篇探讨 AI 时代代码“churn”（频繁重写）观念转变的随笔：当编写代码的成本大幅下降，可抛弃的高质量“一年生”代码与稳定的“多年生”核心架构应被明确区分，churn 本身不是问题，问题在于承重部位的无计划变动。
A practitioner shares a pattern where an AI agent's tool call returns success (200/id) but the intended action never actually completed (e.g., email stuck in drafts). They propose read-back verification after every write, labeling actions as verified, acknowledged, or contradicted, and ask how others handle agent action verification.
A solo founder had Claude Code audit the backend of their AI-built app and discovered six silent bugs—including broken rate limiting, a crashable async error handler, and a Stripe webhook that could lose paid orders. They share lessons on verifying AI-generated code and scheduling regular audits.
An explanation of Change Coupling — how code that must change together even when structurally unrelated creates hidden maintenance costs — and how it relates to other types of coupling as a sense-making tool for architectural decisions rather than a metric to optimize.
Analyzes a new AI development workflow shared by Anthropic employee Thariq, highlighting how replacing Markdown with HTML and SVG can dramatically improve multi-agent collaboration and interaction efficiency, offering a model better suited to human-AI synergy in the AI era.