I started using Codex to learn DevOps… now I’m wondering what exactly I’m learning 😂
一位质量保证工程师正在通过构建个人项目来学习DevOps，从而获得Git、TypeScript、Playwright、测试、CI/CD和架构方面的实践经验。他们正在使用AI工具Codex，并对自己的学习内容产生了疑问。他们目前的策略是手动逐步构建新概念，同时主要利用AI代理进行审查、重复性工作和错误捕捉，并对其他人如何处理这种学习方法感到好奇。
为什么是这条这篇帖子独特地展示了开发者在平衡AI辅助与实际技能习得方面的真实困境，不同于通常只关注AI利弊的报道。
时间与来源
时间显示为 UTC
显示时区：UTC
本地时区尚不可用，暂时显示 UTC。
发布当时偏移：UTC+02026年9月8日 18:38 UTC
收录当时偏移：UTC+02026年9月9日 17:00 UTC
- 发布
- 2026年9月8日 18:38
- 收录
- 2026年9月9日 17:00
- 来源类型
- 开发者社区
- 档位
- 社区
- 信源状态
- 正常
档位是按信源手工设定的编辑判断，不是逐条打分。
讨论趋势
百分比基于采集到的讨论信号，不代表新增评论数或独立参与人数。曲线仅用于同一话题在不同时段的比较。
I’m a QA engineer trying to move into DevOps, so I started building my own project to get hands-on experience with Git, TypeScript, Playwright, testing, CI/CD, and architecture.
Today I decided to try Codex. I gave it access to my local repo and explained what needed to be done.
The thing inspected the code, found the problem, modified the files, added regression tests, cleaned up Git line-ending issues, ran all validations, and created the commit—while I continued doing my actual job.
Then it just came back with a clean summary of everything it had completed.
Absolutely incredible… but also: what the hell am I supposed to learn now? 😂
For those learning DevOps or software engineering while using coding agents: how do you balance the productivity boost without outsourcing the entire learning process?
My current idea is to build new concepts manually, step by step, and use agents mainly for reviews, repetitive work, and catching mistakes. Curious how others approach this.