以下文章来源于伯乐在线 ，作者伯小乐 伯乐在线 . 伯乐在线分享IT互联网职场和精选干货文章（原域名已不再维护）。组织维护10万+star的开源技术资源 库，包括：Python, Java, C/C++, Go, JS, CSS, Node.js, PHP, .NET 等 【导读】：Armin Ronacher 可是编程圈响当当的技术大佬，如果你是 Python 开发者，大概率早有耳闻！他 不仅是资深开源爱好者，更一手打造了 Flask、Jinja2、Sphinx 等知名工具。 智能体元年 在 2025 年收尾、2026 年将至之际，这位大佬写下这篇长文，毫无保留地分享了自己用 AI 编程一整年的真实 体验 —— 从工具选型、工作模式变革，到与机器的相处困惑，再到对行业未来的思考，全是干货满满的心得 体会。 （本文有点长，将近 5000 字，赶时间的盆友，可以先收藏后再细看 ） 氛围编程一年有感 2025 年即将落幕，这一年过得可太不一般了。去年这个时候，我写了篇文章复盘自己的生活。要是当时写的 是编程相关的内容，现在估计早就过时了，毕竟 2025 年对我的职业来说，是前所未有的一年。 不一样的 2025 年 ...

Core Insights - The article discusses the evolution of software engineering towards "Vibe Coding" and the necessity for engineers to adapt to AI-driven development methods, emphasizing that traditional coding practices are becoming obsolete [2][3][4]. Group 1: Steve Yegge's Career and Contributions - Steve Yegge has over 30 years of experience in software development, having worked at Amazon and Google, where he played a crucial role in building technical infrastructures and developing tools like Grok [2][3]. - After leaving Google in 2018 due to perceived conservatism, Yegge joined Grab and later Sourcegraph, where he led the company's transition towards AI-driven development [3][4]. Group 2: Vibe Coding and AI Programming - Yegge argues that using traditional IDEs for coding is no longer acceptable for competent engineers, who must transition to agent programming, where the focus is on managing AI agents rather than writing code directly [5][6][9]. - He emphasizes that the core skill has shifted from coding to directing AI agents, and that engineers who do not embrace AI will quickly fall behind [10][11]. Group 3: Challenges and Future of Software Development - The article highlights the challenges of code merging in high-productivity environments, where traditional methods are insufficient to handle the volume of code changes [33][34]. - Yegge predicts that the future of programming will involve a shift towards "factory-style coding," where AI tools will automate much of the coding process, fundamentally changing team structures and workflows [38][39]. Group 4: Current State of AI Companies - Yegge notes that companies like Google, Anthropic, and OpenAI are currently experiencing internal chaos due to rapid expansion and the challenges of integrating AI into their workflows [45][46]. - He suggests that while these companies are making progress, they still face significant execution challenges that need to be addressed for successful AI integration [47][48].

这家总部位于斯德哥尔摩的公司于两周前宣布，已完成3.3亿美元新一轮融资。本轮融资由CapitalG与Menlo Ventures领投，Khosla Ventures、Accel、殷拓 集团（EQT）等多家机构参与投资。此轮融资使这家初创企业的融资总额达到5.5亿美元。 据《福布斯》估算，35岁的安东·奥西卡与26岁的法比安·赫丁各自持有公司约24%的股份，二人身家均达16亿美元。26岁的赫丁由此成为欧洲最年轻的白 手起家亿万富豪之一，仅比另一位瑞典人工智能研究者——AI编程工具Cursor联合创始人阿维德·伦内马克（Arvid Lunnemark）晚一个月；伦内马克于今 年11月以26岁之龄跻身亿万富豪行列。目前，赫丁已跻身全球仅有的13位30岁以下白手起家亿万富豪之列。 两位创始人均计划，未来无论以何种方式退出，他们都将把收益的50%捐赠给慈善事业。 "我和法比安已经作出承诺，要从我们通过Lovable创造的财富中，拿出很大一部分用于确保人类向超级智能AI平稳过渡，让这项技术惠及每一个身处这个 时代的人。"奥西卡在今年7月接受《福布斯》采访时表示。 Lovable推出的AI编程工具，能让用户仅凭文字指令就能 ...

以下文章来源于快鲤鱼 ，作者好好 快鲤鱼 . 创业邦旗下AGI矩阵号，寻找海内外创新性的AGI高成长公司，记录AGI商业领袖的成长轨迹。 作者丨 好好 编辑丨 小龙 图源 丨 Lovable 12月18日，瑞典AI初创公司Lovable宣布完成B轮融资，本轮融资共获得3.3亿美元资金（约合人民币23亿元），估值达到了66亿美元（约合人民币约464 亿元）。本轮融资由谷歌旗下CapitalG和门罗风险投资共同领投，英伟达风投部门NVentures、阿克塞尔投资、科斯拉风投等多家知名机构参投。 Lovable创始人之一Anton Osika是位90后。他12岁开始编程，拥有欧洲核子研究组织（CERN）粒子物理研究的背景，此后在音乐流媒体巨头Spotify担 任高级机器学习工程师。创立Lovable之前，他曾是AI独角兽Sana Labs的首位工程师，完成一款AI学习平台从0到1的产品技术搭建，是奠定产品技术基因 的关键人物。 Anton Osika 而后，Osika作为CTO创立了电商AI公司Depict.ai，这家公司后来成功入驻了创业孵化器Y Combinator，并累计筹集了超过2000万美元的风险投资获得 ...

又一家大厂，建起了自己的代码防火墙。 文 | 超聚焦 近日，据鞭牛士报道，相关人士爆料称，快手的研发线发布通知，对几款第三方编程软件收紧了使用权 限。不少同学发现，只要在自己办公电脑上点开Cursor，它就直接闪退，根本用不了。 这导致一些早已将AI编程工具深度融入日常开发流程的工程师措手不及。主力辅助工具的突然失效， 不仅打断了高度自动化的编码节奏，也让许多原本由AI即时补全或生成的环节被迫回归手动操作，整 体开发效率明显下滑。 原本几分钟就能自动生成的模板代码，现在又得手动敲；原本靠自然语言描述就能完成的函数逻辑，如 今不得不重新翻文档、查API。效率断崖式下跌，工作流被打断，甚至员工调侃："没Cursor的日子，我 连Hello World都写不顺了。" 而这其实并非孤例。在过去两年中AI工具从"新奇玩具"迅速演变为"开发标配"的过程中，越来越多技术 团队开始面临一个两难命题：一边是肉眼可见的效率跃升，一边是难以量化的安全隐忧。 当一行由远程模型生成的代码可能携带着未知的数据回传、训练偏见甚至知识产权瑕疵，企业不得不重 新评估——所谓"智能辅助"，边界究竟该划在哪里？ 那么，类似的情况在其他大厂是否也已 ...

"氛围编程"可以说是今年最热最出圈的话题了。然而，从"用 LLM 快速拼装应用"爆红，到现在也不过才六个月，就明显开始"退潮"了。 最直观的体现，是全线产品的用户流量出现大幅下滑。 降幅最大的当属 Lovable，其流量在近几个月内从 3500 万掉到不足 2000 万，几乎砍半。其它几个明星产品也没好到哪里去，Bolt.new 下降了 27%，而 Vercel v0 自 5 月以来下降了 64%，Cursor、Replit、Devin 等平台也未能独善其身，唯一例外的是仍在依靠投放驱动的 Base44。 连 Bolt.new 的 CEO 也公开承认，"所有平台的用户流失率都非常高"，并表示当务之急是构建能留住用户的业务模式。 过去一年，行业经历了资本驱动的"超高速增长期"，公司估值与用户数同步飙升；然而当热度在短短几个月内迅速回落，我们或许正在见证一场真正的价 值回归。 | 12wk Change | ela | 5/23 | 616 | 6/20 | 714 | 7/18 | 8/1 | 8/15 | 8/29 | 9/12 | 9/26 | 10/10 | | --- | --- | --- | --- ...

Core Insights - Google recently launched Antigravity, a new IDE touted as the "next-generation agentic development platform," which aims to streamline the entire development process through AI integration. However, early users reported significant issues, including task interruptions due to "model overload" and rapid depletion of credit limits, leading to a poor initial experience [2][26][33] - There are indications that Antigravity is not as original as claimed, with many developers suggesting it is a proprietary fork of Windsurf, a closed-source IDE for which Google paid approximately $2.4 billion for technology licensing [4][6][19] Development and Technical Aspects - The term "PORK" (Proprietary Fork) has been introduced to describe Google's action of forking a closed-source software, which differs significantly from traditional open-source forks in terms of licensing and transparency [4][6] - The similarities between Antigravity and Windsurf are striking, with many UI elements and functionalities appearing almost identical, leading to speculation that Google did not significantly modify the underlying code [7][9][19] - Some developers have noted that the internal structure and naming conventions within Antigravity closely mirror those of Windsurf, suggesting a lack of substantial rework [9][13] Market Reactions and Community Feedback - The launch of Antigravity has sparked discussions in the developer community, with many users humorously comparing it to "copying homework" due to its apparent similarities to Windsurf [16][19] - Despite the ambitious vision for Antigravity as a platform that emphasizes agent-driven development, the initial user experience has been marred by technical issues and a lack of essential features [26][33] Future Vision and Strategic Direction - The founder of Antigravity, Varun, has articulated a vision where the platform is not merely an enhancement of existing IDEs like Cursor or Windsurf but represents a paradigm shift towards an agent-centric development ecosystem [21][23] - Antigravity is designed to allow developers to orchestrate multiple agents simultaneously, marking a departure from the traditional single-agent model, which could significantly change the workflow in software development [22][23] Security and Reliability Concerns - There are ongoing concerns regarding the security and reliability of Antigravity, with warnings about potential data leaks and the need for careful validation of agent actions [34][35] - The rapid development and deployment of Antigravity, following the acquisition of Windsurf's team, raises questions about the thoroughness of testing and the readiness of the product for widespread use [26][34]

鹭羽 发自 凹非寺 量子位 | 公众号 QbitAI 没有logo、没有招聘JD、没有万恶之源KPI。 反之，全员主动996，写代码互相来找茬，甚至公司最有压力的不是程序员，而是厨子…… 我重生了，重生在 Cursor 的"草根"时期。 成立不到两年，估值就超百亿美元，一经推出全网引爆"氛围编程"，不只是让写代码更快，更是重新定义了写代码这事。 | 构建软件的新方式。 | | | | --- | --- | --- | | 前后两批次的效果判若云泥,采用率从个位数飙升至 | 迄今为止我付费使用、毫无疑问最有用的 AI 工具就是 | 最出色的 LLM 应用都有一个"自主性滑杆":你可以决 | | 80%以上。它像野火般迅速蔓延,最顶尖的开发者都 | Cursor。它速度快、在你需要的时机和位置智能补 | 定给 Al 多大的自主权。在 Cursor 中,你可以用 Tab 自 | | 在使用 Cursor。 | 全,括号处理得当,键盘快捷键设计合理,支持自带 | 动补全、用 Cmd+K 做定向编辑,或者直接放手交给全 | | 模型 各方面都打磨得非常到位。 | | 自主代理模式来处理。 | | Diana Hu 胡 ...

Core Insights - The article questions the effectiveness of AI in enhancing software development productivity, highlighting a discrepancy between developers' perceptions and actual performance metrics [1][6][21] - Despite widespread adoption of AI tools, there is no significant increase in the volume of new software releases, contradicting claims of enhanced productivity [8][20][21] Group 1: AI Productivity Claims - Developers believe AI tools improve their productivity by approximately 20%, but research indicates a decline in actual speed by 19% [1][6] - A personal experiment showed that using AI tools resulted in a median speed decrease of 21%, aligning with the findings of the METR report [4][21] - Prominent AI tools like GitHub Copilot and Claude Code claim to significantly boost productivity, yet these claims lack substantial evidence [6][7] Group 2: Software Release Trends - Data analysis reveals no significant increase in new software releases, suggesting that the anticipated "flood" of applications has not materialized [8][20] - The expected exponential growth in software production due to AI adoption is not reflected in the data, which shows flat trends instead [20][21] - The core premise that AI tools would lead to increased output is fundamentally flawed, as developers are not delivering more products than before [21][23] Group 3: Industry Implications - The pressure on developers to adopt AI tools is driven by a fear of missing out on competitive advantages, leading to organizational restructuring and layoffs [7][21] - Many developers feel overwhelmed by the need to master AI tools, which can lead to decreased job satisfaction and productivity [21][22] - The narrative surrounding AI's transformative impact on software development is challenged by the lack of tangible results and evidence of improved efficiency [22][23]

Core Insights - The article discusses the impact of AI programming assistants on software development, highlighting that while they can generate code quickly, they do not replace the need for understanding requirements, designing architecture, and testing [1][2][3] Group 1: AI Programming Assistants - AI programming assistants like Claude Code can generate code rapidly, but they often lead to increased time spent on understanding and integrating the code due to lack of context awareness [2][3] - The efficiency gains from AI coding are often overstated, with real-world productivity improvements being around 10% rather than the claimed 10-fold increase [3][5] Group 2: Challenges in Software Development - Developers are increasingly spending time on "cleanup" tasks such as testing, documentation, and deployment, rather than actual coding [5][6] - The phenomenon of "technical lead's dilemma" arises as experienced engineers face the challenge of balancing team development and project delivery speed [6][9] Group 3: Team Dynamics and Leadership - Technical leads often have to choose between equitable task distribution for team growth or taking on critical tasks themselves to ensure faster delivery, which can harm long-term team development [7][9] - A balanced approach is necessary to ensure both project delivery and team growth, emphasizing the importance of structured practices [11][12] Group 4: Best Practices for AI Integration - Effective integration of AI in software development requires adherence to best practices such as code reviews, modular design, and test-driven development (TDD) [15][23] - AI can assist in various stages of the development lifecycle, including requirement analysis and documentation generation, but must be guided by human oversight to ensure quality and maintainability [20][23] Group 5: Avoiding AI Programming Pitfalls - Companies must establish new engineering practices to effectively leverage AI tools, ensuring that the speed of AI-generated code translates into sustainable delivery capabilities [18][21] - By treating AI as a "lightning-fast junior engineer," organizations can apply historical best practices to build scalable systems [18][21]