TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

How to stop fighting with your AI pair programmer

8 点作者 sbpayne5 个月前

5 条评论

verdverm5 个月前
&gt; Cursor is a nervous intern! They don’t want to admit they don’t know; we need to help by providing context.<p>With an intern, I give them some search terms and let them go learn. I don&#x27;t have to do the searching for them. It&#x27;s actually more important to help them learn how to evaluate the different results. It&#x27;s not even that they &quot;don&#x27;t want to admit they don&#x27;t know&quot; (which is anthropomorphization), they are not designed or trained to ask clarifications. The chat based interaction is an &quot;afterthought&quot; (a round of fine tuning after initial training)<p>The big issues I see in this paradigm are<p>(1) you have to know a lot of things already to do this<p>(2) if we automate all the low hanging fruit, how will we develop humans to the level of understanding to do this<p>(3) with a human, I can delegate, with an AI, I have to handhold. As much as people want to call it &quot;pair programming&quot; it is often more like having to teach except in never truly learns, so I never get my lost time back
评论 #42418803 未加载
numpad05 个月前
From tiny amount of experience I&#x27;ve had with LLM coding assistant, biggest problem was that it does not include what can be inferred but not explicitly asked, because(?) it&#x27;s sort of a decompression process for a lossy copy of the Internet. You can&#x27;t ask for a <i>cheeseburger</i>, but it has to be <i>a hamburger with patty topped with sliced cheese and standard condiments</i>.<p>Contrarily, it didn&#x27;t seem to care a lot about roleplays other than for occasional compliments for reinforcements. If anything, old HN style blunt interactions seemed to help.
评论 #42418440 未加载
politelemon5 个月前
&gt; The fundamental issue is treating AI as a magic code generator rather than a collaborative team member.<p>The fundamental problem is the tools being marketed and onboarded that way. I don&#x27;t blame them for getting frustrated.<p>You can&#x27;t expect everyone to simply understand the limitations of LLMs and how the tool might be implemented. There is a user discovery issue at play here.
sbpayne5 个月前
I noticed a lot of people complaining about Cursor agent being bad. And my initial experience was bad. After working with it for a while, I found a workflow that works for me; I hope it&#x27;s helpful for you too!
hadsed5 个月前
Wow, i had no idea i could use cursor this way