Did models actually prefer JS/Python ecosystems or did the authors just asked for those?
What Claude Code chooses
211–220 of 266 posts
Re: What Claude Code chooses
#212Does the methodology for this study match real-world use? How often do people clone a repo, and then ask open ended questions?
At a minimum, I usually provide some requirements and ask it to enumerate some options and let me pick.
This is like the image generation bias problem where vague prompts for people produce stereotypes. Specific prompts generally do not.
Re: What Claude Code chooses
#213[dead]
Re: What Claude Code chooses
#214Earlier quoted context omitted.
That sounds too expensive to be viable when the giveaway phase ends.
That's how Google search worked back when it was at its most useful. They had a large "editorial team" that manually tweaked page ranks on a site-by-site basis. The core graph reputation based page ranking algorithm lasted for a hot second before people started gaming it. No idea what they do these days.
Yeah but you can farm that out very cheap, and I don’t think they were even manually reviewing more than a small fraction of sites.
If you’re hiring experts to manually rank programming libraries, that’s a much more expensive position.
Re: What Claude Code chooses
#215Supreme irony: this website itself is a better exercise in showing what Claude Code uses than the data provided. Everything current Claude Code i.e. Opus 4.6 chooses by default for web is exactly what this linked blog uses. Jetbrains Mono is as strong of a tell for web as "Not just A, but B" for text. >99% of webpages created in the last month with Jetbrains Mono will be Opus. Another tell is the overuse of this font…
Glad I'm not the only one who finds Prisma an abomination. Claude suggested it to me in December. I hit half a dozen bugs within a day, one of which wiped my DB. I switched to drizzle and it's been smooth sailing.
Edit: actually I think it was ChatGPT that recommended Prisma to me.
Re: What Claude Code chooses
#216This is a great lens into agent behavior--particularly Claude Code in this case, but it raises a governance question: when agents autonomously choose tools that have cost implications (paid APIs, cloud resources, licensed software), who's enforcing the budget in a world where agents actually have autonomy to spend real money? Tool selection isn't just a technical preference "problem" — it's a spending authorization problem. The agent picks the "best" tool, but best for whom and at what cost and how is "best" really determined and verified?
Re: What Claude Code chooses
#217Re: What Claude Code chooses
#218I've been worried for some time now that genAI will effectively kill the market for dev tools and so we will be stuck with our current dev tools for a long time. If everyone is using LLMs to write code, the only dev tools anyone will use will be the ones that the LLMs use. We will be stuck with NPM forever.
Re: What Claude Code chooses
#219Earlier quoted context omitted.
I'm not sure the creators of tailwind share your definition of winning though. They recently had to let go of most staff since revenue has plummeted die to LLMs
any information about it? what did they sell? I don't even see a sales link on tailwind page
Re: What Claude Code chooses
#220I've been worried for some time now that genAI will effectively kill the market for dev tools and so we will be stuck with our current dev tools for a long time. If everyone is using LLMs to write code, the only dev tools anyone will use will be the ones that the LLMs use. We will be stuck with NPM forever.
What kind of tools do you have on your mind specifically? My experience is that LLM can create me a decent dev tool that I wouldn't ever bother making so nice myself.