ทันเอไอ

Coding with AI is no longer about typing code, it's about "being a manager"

A widely discussed article on Hacker News points out that the developer's role is shifting from writing code to overseeing, reviewing, and making decisions for AI to execute.

Allen Bargi's article "Working with AI feels more like leadership than coding" has become a major topic of discussion on Hacker News, garnering 251 points and over 166 comments. The core message is that AI-driven software development is transforming the developer's role from an Individual Contributor into an overseer or leader.

The author explains that instead of focusing on syntax or manual coding, the key tasks now involve defining system architectures, setting clear goals, conducting code reviews, and making strategic decisions to ensure AI delivers the desired output—resembling team management rather than hands-on execution. Consequently, the crucial skills are shifting toward communication, verification, and engineering judgment, replacing the "typing speed" once used to measure a programmer's prowess in the past.

This concept aligns with insights from other experts. InfoQ reported an interview with Alasdair Allan, who noted that programming is becoming increasingly supervisory, especially when dealing with AI-generated code that often presents complexity and quality assurance challenges. Meanwhile, an article by Ashish Sharda on Medium reinforced the same idea: software engineers are moving away from measuring output by lines of code to evaluating their ability to review and decide which code snippets should be deployed into production.

In the Hacker News thread itself, many developers shared firsthand experiences of how their daily routines have shifted to reading, reviewing, and fixing AI-generated work rather than writing from scratch. While this perspective remains an opinion piece based on personal experience rather than industry-wide statistics, the broad discussion reflects that this issue deeply resonates with industry professionals.