Notes on rolling out Cursor and Claude Code

&gt; So far the biggest limiting factor is remembering to use it. Even people I consider power users (based on their Claude token usage) agree with the sentiment that sometimes you just forget to ask Claude to do a task for you, and end up doing it manually. Sometimes you only notice that Claude could have done it, once you are finished. This happens to me an embarrassing amount.<p>Yea, this happens to me too. Does it say something about the tool?<p>It&#x27;s not like we are talking about luddites who refuse to adopt the technology, but rather a group who is very open to use it. And yet sometimes, we &quot;forget&quot;.<p>I very rarely regret forgetting. I feel a combination of (a) it&#x27;s good practice, I don&#x27;t want my skills to wither and (b) I don&#x27;t think the AI would&#x27;ve been <i>that</i> much faster, considering the cost of thinking the prompt and that I was probably in flow.

&gt;Our head of product is a reformed lawyer who taught himself to code while working here. He’s shipped 150 PRs in the last 12 months.<p>&gt;The product manager he sits next to has shipped 130 PRs in the last 12 months.<p>In a serious organization, non technical people should not be shipping any sort of code. They should be doing the highest leverage things possible to help the business, and if that seems to be coding, there are grave issues in the company.

I&#x27;m still having a hard time with coding agents. They are useful but also somehow immature hence dangerous. The other day I asked copilot with GPT4o to add docstrings to my functions in a long Python file. It did a good job on the first half. But when I looked carefully, I realized the second half of my file was gone. Just like that. Half of my file had been silently deleted, replaced by a single terrifying comment along the lines of &quot;continue similarly with the rest of the file&quot;. I use Git of course so I could recover my deleted code. But I feel I still can&#x27;t fully trust an AI assistant that will silently delete hundreds of lines of my codebase just because it is too lazy or something.

&gt; The most common thing that makes agentic code ugly is the overuse of comments.<p>I&#x27;ve seen this complaint a lot, and I honestly don&#x27;t get it. I have a feeling it helps LLMs write better code. And removing comments can be done in the reading pass, somewhat forcing you to go through the code line by line and &quot;accept&quot; the code that way. In the grand scheme of things, if this were the only downside to using LLM-based coding agents, I think we&#x27;ve come a long way.

I have a confession. I don’t really get what Claude Code is… It’s not a model, it’s not an editor with AI integrated… So what is it? It bugs me on the website, I click on it, read, still don’t get it.<p>I have a Claude console account, if you can call it that? It always takes me 3 times to get the correct email address because it does not work with passkeys or anything that lets me store credentials. I just added the api key to OpenWebUI. It’s nice and cheaper than a subscription for me even though I use it all day.<p>But I’m still confused. I just now clicked on “build with Claude”, it takes me to that page where I put in the wrong email address 3 times. And then you can buy credits.

&gt; The product manager he sits next to has shipped 130 PRs in the last 12 months. When we look for easy wins and small tasks for new starters, it’s harder now, because he’s always got an agent chewing through those in the background.<p>I&#x27;d be curious to hear more about this, whether from the author or from someone who does something similar. When the author says &quot;background&quot;, does that literally mean JIRA tickets are being assigned to the agent, and it&#x27;s spitting back full PRs? Is this setup practical?

&gt; You can see this in practice when you use Claude Code, which is pay-per-token. Our heaviest users are using $50&#x2F;month of tokens. That’s a lot of tokens.<p>How is your usage so low! Every time i do anything with claude code i spend couple of bucks, for a day of coding it&#x27;s about $20. Is there a way to save on tokens on a mid-sized Python project or people are just using it less?

&quot;Making it easy to run tests with a single command. We used to do development &amp; run tests via docker over ssh. It was a good idea at the time. But fixing a few things so that we could run tests locally meant we could ask the agent to run (and fix!) tests after writing code.&quot;<p>Good devops practices make AI coding easier!

Changing my old coding behavior aside, biggest limiting factor for me is understanding how and why the coding agent is doing this a certain way, so that I have the confidence to continually sharpen my tools.<p>I want something simple that I have full control on, if not just to understand how they work. So I made a minimal coding agent (with edit capability) that is fully functional using only seven tools: read, write, diff, browse, command, ask, and think.<p>As an example, I can just disable `ask` tool to have it easily go full autonomous on certain tasks. Or, ask it to `think` for refactoring.<p>Have a look at <a href="https:&#x2F;&#x2F;github.com&#x2F;aperoc&#x2F;toolkami">https:&#x2F;&#x2F;github.com&#x2F;aperoc&#x2F;toolkami</a> to see if it might be useful for you.

Good to see experiences from people rolling out AI code assistance at scale. For me the part that resonates the most is the ambition unlock. Using Brokk to build Brokk (a new kind of code assistant focused on supervising AI rather than autocompletes, <a href="https:&#x2F;&#x2F;brokk.ai&#x2F;" rel="nofollow">https:&#x2F;&#x2F;brokk.ai&#x2F;</a>) I&#x27;m seriously considering writing my own type inference engine for dynamic languages which would have been unthinkable even a year ago. (But for now, Brokk is using Joern with a side helping of tree-sitter.)

Having linting&#x2F;prettifying and fast test runs in Cursor is absolutely necessary. On a new-ish React Typescript project, all the frontier models insist on using outdated React patterns which consistently need to be corrected after every generation.<p>Now I only wish for an Product Manager model that can render the code and provide feedback on the UI issues. Using Cursor and Gemini, we were able to get a impressively polished UI, but it needed a lot of guidance.<p>&gt; I haven’t yet come across an agent that can write beautiful code.<p>Yes, the AI don&#x27;t mind hundreds of lines of if statements, as long as it works it&#x27;s happy. It&#x27;s another thing that needs several rounds of feedback and adjustments to make it human-friendly. I guess you could argue that human-friendly code is soon a thing of the past, so maybe there&#x27;s no point fixing that part.<p>I think improving the feedback loops and reducing the frequency of &quot;obvious&quot; issues would do a lot to increase the one-shot quality and raise the productivity gains even further.

Author here. Ask me anything!<p>Since writing this a tangentially related thing we&#x27;ve added, is a github action that runs on any PR that includes a (Rails) database migration, and reviews it, comparing it to our docs for how to write good migrations.<p>Claude helped write the action so it was super easy to set up.

&gt; You can see this in practice when you use Claude Code, which is pay-per-token. Our heaviest users are using $50&#x2F;month of tokens. That’s a lot of tokens. I asked our CFO and he said he’d be happy to spend $100&#x2F;dev&#x2F;month on agents. To get 20% more productive that’s a bargain.<p>fwiw we interviewd the Claude Code team (<a href="https:&#x2F;&#x2F;www.latent.space&#x2F;p&#x2F;claude-code" rel="nofollow">https:&#x2F;&#x2F;www.latent.space&#x2F;p&#x2F;claude-code</a>) and they said that even within Anthropic (where Claude is free, we got into this a bit), the usage is $6&#x2F;day so about $200&#x2F;month. not bad! especially because it goes down when you under-use.

&gt;Our head of product is a reformed lawyer who taught himself to code while working here. He’s shipped 150 PRs in the last 12 months.<p>&gt;The product manager he sits next to has shipped 130 PRs in the last 12 months.<p>this is actually horrifying, lol. i haven&#x27;t even considered product guys going ham on the codebase

&gt; I haven’t yet come across an agent that can write beautiful code.<p>o3 in codex is pretty close sometimes. I prefer to use it for planning&#x2F;review but it far exceeds my expectations (and sometimes my own abilities) quite regularly.

As someone who really dislikes using Cursor, what does the HN hivemind think of alternatives? Is there a good CLI like Claude Code but for Gemini &#x2F; other models? Is there a good Neovim plugin that gets the contextual agent mode right?

Even if you don&#x27;t think AI will replace the job of software developer completely, there&#x27;s no way compensation is going to stay at the current level when anyone can ship code.

I&#x27;m wondering how much better llms are at &quot;untyped&quot;&#x2F;loosely typed scripters like Ruby for large scale coding and error avoidance.<p>Presumably an llm can actually maintain better contextual awareness of code and variables than, say cold loaded syntax highlights.