Limboy

Vibe code is legacy code

Updates and articles from the Val Town team

blog.val.town

完全同意文章的观点:将 Vibe Coding 比作给孩子一张信用卡,初期会带来极大的满足感,但最终会面临无法偿还的“债务”——即无法理解和维护的代码。编程的本质是“理论构建”,而非简单地生成代码行。在 AI 时代,开发者仍需保持对代码的理解和控制,像对待一个“过度热情但爱胡说八道的初级实习生”一样对待AI。

Vibe Coding 适合构建一次性工具和原型,对于需要持续维护(尤其是有较高复杂度)的项目,一定要谨慎 Vibe。

肝了6个月的AudioLLM,开源了【100亿模型计划】

代码: https://github.com/boson-ai/higgs-audio

www.youtube.com

李沐的讲解视频真不错,一些复杂的概念经他一拆解,就变得容易理解了。

因喜欢而去做还是因为做而喜欢?

我们是因喜欢而去做一件事,还是因去做一件事而慢慢喜欢?这个问题不仅是一个简单的先后顺序之辩,更是贯穿人类经验的核心张力。它将“追随激情”的浪漫主义理想与“爱上你所做之事”的务实主义现实置于对立面。本报告旨在剖析这两种路径,揭示它们并非相互排斥的选项,而是一个动态、互联系统中的两端。

limboy.me

中年转型期: 通常以对过往道路的质疑为特征 。在获得了一定程度的胜任感和安全感之后,对自主性和意义的需求变得更加突出。这可能导致“中年危机”,或者更具建设性地,促成一次转向更具激情驱动的职业变革。许多关于在人生后半段找到激情的故事都符合这一模式 。

看来我已经进入了「中年转型期」···

“追随你的激情”这一简单化的建议往往存在缺陷,因为激情很少是一个预先存在、等待被发现的实体。它通常是发展技能的结果,而非原因 。一种更有效的方法是“工匠思维”(Craftsman Mindset):专注于发展有价值的技能(“让自己优秀到无法被忽视”)。这种技能的精通能够建立胜任感,进而培养激情,并提供塑造充实生活所需的自主性和职业资本。

ikigai 框架

  • 你所热爱的(What you LOVE) - 激情
  • 你所擅长的(What you are GOOD AT) - 职业/技能
  • 世界所需要的(What the WORLD NEEDS) - 使命
  • 你能获得报酬的(What you can be PAID FOR) - 专业/现实

Download Clear: Uniquely Charming Lists for iPhone. Now 100% Free!

The new Clear is out. Available on the Apple Appstore now.

www.useclear.com

Was searching apps for inspiration, wondering if Clear is still accessible, and it is! Still updating!

I stumbled upon this blog and saw the latest article, "15 Years Ago Yesterday." The first sentence immediately grabbed my attention:

"Fifteen years ago yesterday, I was 65 and lying on an operating table at UCLA Medical Center, about to have open heart surgery."

It's incredible to think this person is 80 years old and still blogging frequently! That's so cool!

【BACKUP】中島美嘉 曾經我也想過一了百了 僕が死のうと思ったのは Live【中字】

www.youtube.com

曾经我也想过一了百了
是因为还没有遇见你
像你这样的人存在这世界上
让我稍微对这世界感到喜欢

关于 Yak Shaving

antfu.me

Antfu 分享了自己做开源项目的经验和看法,核心观点是“Yak Shaving”(剪牦牛毛)这一概念。Yak Shaving 原本指在完成一项任务时,不断被其他未完成的任务打断,最终偏离目标。然而,作者认为,如果运用得当,Yak Shaving 反而能成为推动进步的重要动力,尤其是在开源领域。

作者通过自己的亲身经历,讲述了如何从解决实际问题出发,逐步孵化出 i18n Ally、VueUse 和 vue-demi 等知名开源项目,并最终加入 Vue 团队,实现全职开源。他强调,成功的开源项目并非仅靠机遇、营销或技术创新,更重要的是开始项目的动机和持续的动力,而这种动力往往来源于解决自己遇到的实际问题。

文章提出了“Shave the Good Yak”的方法论:首先是“发现问题”,即解决自己遇到的实际问题,因为作为用户最了解痛点;其次是“解决问题”,在没有现有解决方案或现有方案不佳时,大胆尝试;接着是“点到为止”,不要抱太大期望,够用即可,及时回到主线任务;然后是“完善项目”,作为第一用户不断改进,并适时分享;最后是“发现更多问题”,通过解决问题和社区反馈进入正循环。

Andrej Karpathy: Software Is Changing (Again)

Andrej Karpathy's keynote on June 17, 2025 at AI Startup School in San Francisco. Slides provided by Andrej: https://drive.google.com/file/d/1a0h1mkwfmV2PlekxDN8isMrDA5evc4wW/view?usp=sharing Chapters (Powered by https://ChapterMe.co) - 00:00 - Intro 01:25 - Software evolution: From 1.0 to 3.0 04:40 - Programming in English: Rise of Software 3.0 06:10 - LLMs as utilities, fabs, and operating systems 11:04 - The new LLM OS and historical computing analogies 14:39 - Psychology of LLMs: People spirits and cognitive quirks 18:22 - Designing LLM apps with partial autonomy 23:40 - The importance of human-AI collaboration loops 26:00 - Lessons from Tesla Autopilot & autonomy sliders 27:52 - The Iron Man analogy: Augmentation vs. agents 29:06 - Vibe Coding: Everyone is now a programmer 33:39 - Building for agents: Future-ready digital infrastructure 38:14 - Summary: We’re in the 1960s of LLMs — time to build Drawing on his work at Stanford, OpenAI, and Tesla, Andrej sees a shift underway. Software is changing, again. We’ve entered the era of “Software 3.0,” where natural language becomes the new programming interface and models do the rest. He explores what this shift means for developers, users, and the design of software itself— that we're not just using new tools, but building a new kind of computer. More content from Andrej: https://www.youtube.com/@AndrejKarpathy Thoughts (From Andrej Karpathy!) 0:49 - Imo fair to say that software is changing quite fundamentally again. LLMs are a new kind of computer, and you program them *in English*. Hence I think they are well deserving of a major version upgrade in terms of software. 6:06 - LLMs have properties of utilities, of fabs, and of operating systems → New LLM OS, fabbed by labs, and distributed like utilities (for now). Many historical analogies apply - imo we are computing circa ~1960s. 14:39 - LLM psychology: LLMs = "people spirits", stochastic simulations of people, where the simulator is an autoregressive Transformer. Since they are trained on human data, they have a kind of emergent psychology, and are simultaneously superhuman in some ways, but also fallible in many others. Given this, how do we productively work with them hand in hand? Switching gears to opportunities... 18:16 - LLMs are "people spirits" → can build partially autonomous products. 29:05 - LLMs are programmed in English → make software highly accessible! (yes, vibe coding) 33:36 - LLMs are new primary consumer/manipulator of digital information (adding to GUIs/humans and APIs/programs) → Build for agents! Some of the links: - Software 2.0 blog post from 2017 https://karpathy.medium.com/software-2-0-a64152b37c35 - How LLMs flip the script on technology diffusion https://karpathy.bearblog.dev/power-to-the-people/ - Vibe coding MenuGen (retrospective) https://karpathy.bearblog.dev/vibe-coding-menugen/

www.youtube.com

本次演讲中,前特斯拉人工智能总监 Andrej Karpathy 探讨了软件的演进,将其划分为三个阶段:软件 1.0(传统代码)、软件 2.0(神经网络)和软件 3.0(大型语言模型)。他强调,大型语言模型(LLM)正以前所未有的方式改变着软件开发,它们可以通过自然语言进行编程,并催生了“部分自主应用”。Karpathy 将 LLM 比作操作系统和公用事业,并指出了其高昂的计算成本和独特的“心理学”特性,如幻觉和顺行性遗忘症。他还分享了利用 LLM 进行快速原型开发的经验,并提出了为 AI 代理构建软件的建议,例如创建 LLM 友好的数据格式和文档。他总结道,未来十年将是部分自主产品蓬勃发展的时代。

TDD, AI agents and coding with Kent Beck

• Kent Beck is one of the most influential figures in modern software development. Creator of Extreme Programming (XP), co-author of The Agile Manifesto, and a pioneer of Test-Driven Development (TDD), he’s shaped how teams write, test, and think about code. Now, with over five decades of programming experience, Kent is still pushing boundaries—this time with AI coding tools. In this episode of Pragmatic Engineer, I sit down with him to talk about what’s changed, what hasn’t, and why he’s more excited than ever to code. In our conversation, we cover: • Why Kent calls AI tools an “unpredictable genie”—and how he’s using them • Why Kent no longer has an emotional attachment to any specific programming language • The backstory of The Agile Manifesto—and why Kent resisted the word “agile” • An overview of XP (Extreme Programming) and how Grady Booch played a role in the name • Tape-to-tape experiments in Kent’s childhood that laid the groundwork for TDD • Kent’s time at Facebook and how he adapted to its culture and use of feature flags • And much more! — Brought to you by: • Sonar — Code quality and code security for ALL code http://sonarsource.com/pragmaticsecurity •⁠ Statsig ⁠ — ⁠ The unified platform for flags, analytics, experiments, and more http://statsig.com/pragmatic • Augment Code — AI coding assistant that pro engineering teams love http://augmentcode.com/pragmatic — The Pragmatic Engineer deepdives relevant for this episode: • Inside Facebook’s engineering culture: https://newsletter.pragmaticengineer.com/p/facebook • Shipping to production: https://newsletter.pragmaticengineer.com/p/shipping-to-production • Software architecture with Grady Booch: https://newsletter.pragmaticengineer.com/p/software-architecture-with-grady-booch • The AI Engineering Stack https://newsletter.pragmaticengineer.com/p/the-ai-engineering-stack • The past and future of modern backend practices https://newsletter.pragmaticengineer.com/p/the-past-and-future-of-backend-practices • AI Engineering in the real world https://newsletter.pragmaticengineer.com/p/ai-engineering-in-the-real-world • Paying down tech debt https://newsletter.pragmaticengineer.com/p/paying-down-tech-debt — Where to find Kent Beck: • X: https://x.com/kentbeck • LinkedIn: https://www.linkedin.com/in/kentbeck/ • Website: https://kentbeck.com/ • Newsletter: https://tidyfirst.substack.com/ Where to find Gergely Orosz: • X: https://x.com/GergelyOrosz • LinkedIn: https://www.linkedin.com/in/gergelyorosz/ • Bluesky: https://bsky.app/profile/gergely.pragmaticengineer.com • Newsletter and blog: https://www.pragmaticengineer.com/ — In this episode, we cover: (00:00) Intro (02:27) What Kent has been up to since writing Tidy First (06:05) Why AI tools are making coding more fun for Kent and why he compares it to a genie (13:41) Why Kent says languages don’t matter anymore (16:56) Kent’s current project building a small talk server (17:51) How Kent got involved with The Agile Manifesto (23:46) Gergely’s time at JP Morgan, and why Kent didn’t like the word ‘agile’ (26:25) An overview of “extreme programming” (XP) (35:41) Kent’s childhood tape-to-tape experiments that inspired TDD (42:11) Kent’s response to Ousterhout’s criticism of TDD (50:05) Why Kent still uses TDD with his AI stack (54:26) How Facebook operated in 2011 (1:04:10) Facebook in 2011 vs. 2017 (1:12:24) Rapid fire round — See the transcript and other references from the episode at https://newsletter.pragmaticengineer.com/podcast — Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [email protected].

www.youtube.com

So you can think really big thoughts and the leverage of having those big thoughts has just suddenly expanded enormously. I had this tweet two years ago where I said "90% of my skills just went to zero dollars and 10% of my skills just went up 1000x". And this is exactly what I'm talking about - having a vision, being able to set milestones towards that vision, keeping track of a design to maintain or control the levels of complexity as you go forward. Those are hugely leveraged skills now compared to knowing where to put the ampersands and the stars and the brackets in Rust.

Gitingest

Replace 'hub' with 'ingest' in any GitHub URL for a prompt-friendly text.

gitingest.com

Turn any Git repository into a simple text digest of its codebase. This is useful for feeding a codebase into any LLM.