Claude (5 blogmarks)

CCC shows that AI systems can internalize the textbook knowledge of a field and apply it coherently at scale. AI can now reliably operate within established engineering practice. This is a genuine milestone that removes much of the drudgery of repetition and allows engineers to start closer to the state of the art.

Right now, where LLM coding agents can do there best work is in codebase where a solid foundation, clear conventions, and robust abstractions and patterns have all been well-established.

As writing code is becoming easier, designing software becomes more important than ever. As custom software becomes cheaper to create, the real challenge becomes choosing the right problems and managing the resulting complexity. I also see big open questions about who is going to maintain all this software.

Managing the complexity of large software systems has always been what makes software engineering challenging. While LLMs handle the “drudgery” of actually writing the code, are they simultaneously exacerbating the challenge of managing the complexity? If so, how do we, as “conductors” of coding agents, reign in that expanding complexity?

Simon Willison shared this addition to Claude’s system prompts:

If the person is unnecessarily rude, mean, or insulting to Claude, Claude doesn't need to apologize and can insist on kindness and dignity from the person it’s talking with. Even if someone is frustrated or unhappy, Claude is deserving of respectful engagement.

It’s very interesting to read through the full prompts to see all the different kinds of instruction they start with.

Just In Time manual reading

The world is full of manuals, how can I decide which ones I should read? One way is to read them all, which used to work back when I was younger, but now I have more responsibilities and I don't have time to read the entire Python library reference including all the useless bits I will never use. But here's Claude pointing out to me that this is something I need to know about, now, today, and I should read this one. That is valuable knowledge.

📦 Repomix (formerly Repopack) is a powerful tool that packs your entire repository into a single, AI-friendly file. Perfect for when you need to feed your codebase to Large Language Models (LLMs) or other AI tools like Claude, ChatGPT, DeepSeek, Perplexity, Gemini, Gemma, Llama, Grok, and more.

I used this for the first time to quickly bundle up a Python program into a single file that I could hand to Claude for help with a setup issue.

Assuming I'm already in the directory of the project, I can run:

$ npx repomix

I've been experimenting with mise lately for managing tool versions like node, so I'll use that.

Here I ask mise to run npx repomix in the context of Node.js v23:

$ mise exec node@23 -- npx repomix

It spit out a file called repomix-output.txt.

I wanted to drag that file from Finder into the Claude app, so I then ran:

$ open . -a Finder.app

An AI coding tool that I use directly from the terminal?! 👀

Claude Code is available as a limited research preview, and enables developers to delegate substantial engineering tasks to Claude directly from their terminal.

"thinking tokens"? Does that mean the input and output tokens that are used as part of intermediate, step-by-step "reasoning"?

In both standard and extended thinking modes, Claude 3.7 Sonnet has the same price as its predecessors: $3 per million input tokens and $15 per million output tokens—which includes thinking tokens.

The Claude Code Overview shows how to get started installing and using Claude Code in the terminal.