有观点认为"大型语言模型可能擅长编码，但软件的难点从不在代码本身"以及"编码很容易，难的是弄清楚要编码什么"[1]。对此，一位作者提出强烈反驳，主张"编码不是难点这一说法是对所有程序员的侮辱"[1]。
作者通过多项证据论证了编码工作的复杂性和高专业门槛[1]。程序员的高薪酬和市场需求反映了这一职业的价值，而大量专业著作如《代码整洁之道》《程序员修炼之道》以及遍布全球的大学课程和编程训练营的存在，都印证了编码是一项需要深度技能的工作[1]。
作者同时承认理解用户需求和业务目标同样重要，主张采纳"两者兼顾"的态度[1]。他建议程序员应当既要深入理解系统本身，也要深入理解为什么要构建它，同时不应将理解、判断、同理心和品味外包给人工智能，也不应成为"肉质代理人"[1]。作者呼吁在人工智能革命中，程序员应当适应变化，既要深化技术专长，也要拓展对相邻领域的理解[1]。
A recent article challenges the increasingly common assertion that coding itself is not the difficult aspect of software development, arguing instead that such dismissals undervalue the craft and expertise of programmers [1]. The author confronts statements like "LLMs may be good at coding, but software was never the hard part" and "coding is easy, it's figuring out what to code that's hard," positioning them as fundamentally disrespectful to the profession [1].
To support this position, the author points to multiple indicators that coding demands substantial skill and dedication: programmer salaries and market demand remain high, professional literature on the subject is extensive—citing works such as Clean Code and The Pragmatic Programmer—and formal education pathways including university programs and coding bootcamps continue to proliferate [1]. Rather than accepting a false dichotomy, the author advocates for "why not both," meaning programmers should simultaneously deepen their understanding of technical systems themselves while also grasping the reasoning behind building them in the first place [1].
The article concludes by urging developers not to outsource their capacity for understanding, judgment, empathy, and taste to artificial intelligence, warning against becoming mere "meat intermediaries" in the age of large language models [1].