--- title: Hardly Promethean | daily.dev description: A response to DHH's Rails World keynote and the ensuing debate (via Dan Luu and Varun Gandhi) about whether LLMs have removed the cost bottleneck in software... canonical: https://daily.dev/posts/hardly-promethean-bqxrqau7m twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Hardly Promethean | daily.dev og:description: A response to DHH's Rails World keynote and the ensuing debate (via Dan Luu and Varun Gandhi) about whether LLMs have removed the cost bottleneck in software... og:url: https://daily.dev/posts/hardly-promethean-bqxrqau7m og:image: https://api.daily.dev/og/posts/bQXrqaU7M.png og:image:alt: Hardly Promethean og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Hardly Promethean **[Lobsters](https://daily.dev/sources/lobsters)** · 6 min read · 0 upvotes · 0 comments ## Summary A response to DHH's Rails World keynote and the ensuing debate (via Dan Luu and Varun Gandhi) about whether LLMs have removed the cost bottleneck in software development. The author argues that LLMs shift the bottleneck from writing code to shipping, maintaining, and reviewing it, and that DHH's rising personal tolerance for sloppy agent-generated code doesn't generalize to teams or shared codebases like Rails. The deeper critique: given supposedly unlimited tokens and 1000x productivity, DHH built only things that already existed (a calculator, a video editor, a Linux distro) rather than anything genuinely new for Rails, suggesting the real scarcity isn't cost but ideas. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why couldn't Bun's LLM-assisted fork of Zig with 4x faster compilation be merged back into upstream Zig? The fork could not be upstreamed because the resulting compiler was non-deterministic, and no maintainer wants a compiler that behaves unpredictably across builds. The fork only needs to work reliably for Bun's own use case, not for the broader Zig community, which has different reliability standards for a shared compiler. _Anyone weighing LLM-assisted compiler or toolchain forks against upstream reliability can follow this trade-off discussion on daily.dev._ ### What is Varun Gandhi's main criticism of the argument that LLMs make software performance work cheap enough for everyone? Gandhi argues the cost-reduction logic only holds for experts working on their own projects, like Dan Luu, and doesn't generalize to most companies or programmers. He points out that writing code was never the dominant cost; shipping, maintaining, and avoiding regressions matter more, and rising tolerance for agent-generated slop under asynchronous workflows undermines the claimed gains. _Developers deciding how much to trust LLM performance claims can dig into this debate on daily.dev before rearchitecting their stack._ ## Similar posts on daily.dev - [Look at all the things he's not doing](https://daily.dev/posts/look-at-all-the-things-he-s-not-doing-e6xpi9hak) · RUBYLAND · 0 upvotes · 0 comments - [The LLM Critics Are Right. I Use LLMs Anyway.](https://daily.dev/posts/the-llm-critics-are-right-i-use-llms-anyway--02xhbt7xl) · Hacker News · 4 upvotes · 1 comments - [Fragments: October 4](https://daily.dev/posts/fragments-october-4-lzsrirk3s) · Martin Fowler · 0 upvotes · 0 comments --- Tags: [#rails](https://daily.dev/tags/rails), [#vibe-coding](https://daily.dev/tags/vibe-coding) [View this post on daily.dev](https://daily.dev/posts/hardly-promethean-bqxrqau7m) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Hardly Promethean","url":"https://daily.dev/posts/hardly-promethean-bqxrqau7m","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/hardly-promethean-bqxrqau7m"},"datePublished":"2026-09-30T13:15:02.260Z","dateModified":"2026-09-30T14:19:35.788Z","description":"A response to DHH's Rails World keynote and the ensuing debate (via Dan Luu and Varun Gandhi) about whether LLMs have removed the cost bottleneck in software...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b162780648317c1642a336752754be5c?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b162780648317c1642a336752754be5c?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Lobsters","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Lobsters","logo":"https://media.daily.dev/image/upload/s--tl8v_Fku--/f_auto,t_logo/v1698841318/logos/lobste.jpg","url":"https://daily.dev/sources/lobsters"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/hardly-promethean-bqxrqau7m","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"rails,vibe-coding","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Lobsters","item":"https://daily.dev/sources/lobsters"},{"@type":"ListItem","position":3,"name":"Hardly Promethean"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/hardly-promethean-bqxrqau7m#faq","mainEntity":[{"@type":"Question","name":"Why couldn't Bun's LLM-assisted fork of Zig with 4x faster compilation be merged back into upstream Zig?","acceptedAnswer":{"@type":"Answer","text":"The fork could not be upstreamed because the resulting compiler was non-deterministic, and no maintainer wants a compiler that behaves unpredictably across builds. The fork only needs to work reliably for Bun's own use case, not for the broader Zig community, which has different reliability standards for a shared compiler. Anyone weighing LLM-assisted compiler or toolchain forks against upstream reliability can follow this trade-off discussion on daily.dev."}},{"@type":"Question","name":"What is Varun Gandhi's main criticism of the argument that LLMs make software performance work cheap enough for everyone?","acceptedAnswer":{"@type":"Answer","text":"Gandhi argues the cost-reduction logic only holds for experts working on their own projects, like Dan Luu, and doesn't generalize to most companies or programmers. He points out that writing code was never the dominant cost; shipping, maintaining, and avoiding regressions matter more, and rising tolerance for agent-generated slop under asynchronous workflows undermines the claimed gains. Developers deciding how much to trust LLM performance claims can dig into this debate on daily.dev before rearchitecting their stack."}}]} ```