How Many Tries We Need Using Coding Agents May Say More About Standards Than Competence

The blog post discusses the experiences of developers using AI coding tools like Claude Code, questioning whether the number of attempts required for satisfactory results indicates a developer's competence. It draws parallels with film directors Kubrick and Wood to illustrate the subjective nature of satisfaction and quality in coding. The author explores the notion that the quality standards developers hold may affect their perception of AI coding efficiency rather than their actual skills.