我们是否应该将 ChatGPT Work 重命名为 ChatGPT Vibe？
Should we rename ChatGPT Work to ChatGPT Vibe?FOLLOW BUILDERS / X FEED
最新 X 动态
27 条公开帖子 · 上游更新时间 2026/7/24 15:10:08
从科幻小说到科学现实。如果您想研究一些最酷、最具影响力的技术，请加入我们的团队。
From Science Fiction to Science Reality. Join the team if you want to work on some of the coolest and most impactful technology.自动缩放部署（通常是最昂贵的缩放应用程序）下降了 80%！ https://t.co/5Thud2waXD https://t.co/mUJwAzV2Xu
Autoscale deployments, which is typically the most expensive of scaled apps is down 80%! https://t.co/5Thud2waXD https://t.co/mUJwAzV2Xu我的国际象棋自动研究代理人获得了现代法学硕士微调博士学位。 https://t.co/xw9mvVRhgY
My chess autoresearch agent got a PhD in modern LLM finetuning. https://t.co/xw9mvVRhgY下一步的发展是能够启动多个 ChatGPT 语音线程，这样我就可以让整个团队与我以及彼此交谈。 🔥
The next evolution is being able to spin up multiple ChatGPT Voice threads so I can have a full team talking to me and to each other. 🔥使用 ChatGPT 语音之前和之后 https://t.co/kzNE5odHSy
Before and after with ChatGPT Voice https://t.co/kzNE5odHSy更多反馈： 1. 如果我有一堆线程在运行，它应该让我知道其他线程何时完成工作 2. 中文发音不好听😅
More feedback: 1. It should let me know when the other threads finish working if I have a bunch of threads going 2. The Chinese pronounciation sounds bad 😅顺便说一句，在过去一个月左右的时间里，我一直在测试一个代理 github 克隆，它使用起来非常愉快。感谢平台工作人员，甚至还完成了内置 CI/CD！ 在上线之前，我还必须实现另外 3 个想法（此处未显示）。但如果您想与我一起解决这个问题，那么现在是加入 swyx inc 并影响路线图的好时机
btw ive been dogfooding an agentic github clone over the past month or so and its gotten quite quite enjoyable to use. even complete with built in CI/CD thanks to workers for platforms! there's 3 more ideas i have to implement (not shown here) before this goes live. but if you're looking to hack on this with me it's a really good time to join swyx inc and influence the roadmap思考人工智能的最佳方式是将其视为你已经了解的领域的力量倍增器（或者你想要了解新领域的速度）。第三类——没有现有的判断，没有兴趣开发它——基本上会产生废品，不会推动那么多的经济生产活动。 将会发生的事情是，专家们的技术会越来越好，并且能够做更多的事情。因为他们可以以实际推动高质量输出的方式运用这些工具，在代理离开时引导他们走向正确的方向，并实际上将工作融入到有用的东西中。 代理的专家工程师将完成更有成效的工作，正是因为他们知道如何正确引导代理。与没有设计眼光的人相比，使用人工智能的设计师会产生更好的结果。等等。 由于这种动态，专业化将继续变得重要，甚至随着工具变得更加强大而变得更加重要。因为市场的预期会变得更高。未来，精通任何一门手艺仍然是必要的。
The best way to think about AI is as a force multiplier for the fields you already know about (or for the rate at which you want to learn about a new one). The 3rd category -no existing judgment, no interest in developing it- will basically produce slop and won’t drive that much economically productive activity. What will happen is that the experts will get better and better at their craft and be able to do so much more. Because they can wield these tools in ways that actually drive high quality output, veer agents in the right direction when they go off, and actually incorporate the work into something useful. The expert engineer with agents will do far more productive work, precisely because they know how to steer the agent properly. The designer will product far better outcomes using AI than someone without an eye for design. And so on. As a result of this dynamic, specialization will continue to be important, if not even more so as the tools get more powerful. Because the expectation of the market gets to be much higher. Getting good at any craft will continue to be necessary in the future.伟大的建设者了解人工智能模型的锯齿状前沿。 伟大的领导者了解其人民的锯齿状边界。
Great builders understand the jagged frontier of AI models. Great leaders understand the jagged frontier of their people.Python 代码现在在 Vercel 上的启动速度提高了 2 倍。自动地！ https://t.co/Id8iH1hSwH
Python code now starts 2x faster on Vercel. Automatically! https://t.co/Id8iH1hSwH当创始人为盈利的自举业务筹集资金，而不是为 Neo-lab 的计算消耗数亿美元时，风险投资 https://t.co/YmF9a7BK5p
VCs when a founder is raising for a profitable bootstrapped business instead of burning hundreds of millions on compute for a neo-lab https://t.co/YmF9a7BK5p是时候在旧金山建造住房了 https://t.co/i0YVOsb3Pn
It’s time to build housing in SF https://t.co/i0YVOsb3Pn贾维斯/萨曼莎/TARS/等 尝试一下，在远离键盘的情况下尽最大努力工作。是时候玩得开心了！ 现已在 ChatGPT 桌面应用程序中提供。 https://t.co/nmimzU5Jta
Jarvis / Samantha / TARS / Etc Try it, and do your best work all while being away from that keyboard. Time to have fun! Available in the ChatGPT desktop app now. https://t.co/nmimzU5Jta我认为人们对@poolsideai不够欣赏的一件事是他们不同寻常的开放程度——他们不仅提供了一个出色的小型模型，在编码方面以某种方式击败了@thinkymachines，而且大多数人（如@eliebakouch）都在大声喊出他们的优秀论文，而且他们也是少数真正公开其完整评估数据集的人之一——精美地发布了6个公共基准测试，每次运行4次，每次运行数百次。如果他们奖励黑客，你就可以满足自己。杰出的。
one thing i think people dont appreciate enough about @poolsideai is their unusual degree of openness — not only have they shipped an excellent Small model that somehow beat @thinkymachines at coding, but most people (like @eliebakouch) have been shouting out their excellent papers, but also they're among a rare few to actually expose their full eval dataset as well - beautifully published, with across 6 public benchmarks with 4 runs each and hundreds of turns per run. you can satisfy for yourself if they rewardhack. brilliant.每个套餐的语音模式还支持更多语言，包括西班牙语、法语、印地语和日语。 该更新今天在移动、桌面和网络上以公开测试版的形式推出。下载移动应用程序并点击声波开始对话：https://t.co/hwPB3zlk0w
Voice mode also supports more languages, on every plan, including Spanish, French, Hindi, and Japanese. The update is rolling out today in public beta on mobile, desktop, and web. Download the mobile app and tap the sound wave to start a conversation: https://t.co/hwPB3zlk0w语音对话现在使用更多聊天中的模型，包括 Claude Opus 和 Sonnet。克劳德还可以访问您在对话中连接的工具，例如电子邮件和日历。 https://t.co/452G2ZZY1d
Voice conversations now use more of the models you have in chat, including Claude Opus and Sonnet. Claude can also reach the tools you've connected mid-conversation, like your email and calendar. https://t.co/452G2ZZY1d语音模式现在可以在 Claude 功能更强大的模型上运行，并可以使用您在对话中连接的工具。 用更多语言大声讨论难题。 https://t.co/k0CWAGjLdK
Voice mode now runs on Claude's more capable models and reaches the tools you've connected mid-conversation. Talk through the hard problems out loud, in many more languages. https://t.co/k0CWAGjLdKViktor 能够颠覆代理模式，并通过使用 Replit 赚了很多钱。 然后他想：为什么不将整个事情自动化，而不仅仅是编码部分呢？ 该机构：这只是一个代理循环...... 所以他向我们的团队请求建立一个 MCP，我们建立了一个，他现在建立了他的自治机构。
Viktor was able to disrupt the agency model and make a lot of money by using Replit. Then he figured: Why not automate the entire thing, not just the coding part? The Agency: It's merely an agent loop... So he asked our team for an MCP, and we built one and he's now built his autonomous agency.废除并改革 CEQA，这是邻避者使用和滥用的一种监管工具，用于封锁加州各地的住房 https://t.co/licr3PVeor
Repeal and reform CEQA, the one regulatory tool that is used and abused by NIMBYs to block housing everywhere in California https://t.co/licr3PVeor开放权重模型非常非常重要 https://t.co/IwS4UYG3pD https://t.co/FhgtJzdqUl
Open weight models are very very important https://t.co/IwS4UYG3pD https://t.co/FhgtJzdqUl🔴🔊AI网关不断进步。团队的产品速度不真实。 https://t.co/OWkksdt10W
🔴🔊 AI Gateway keeps getting better. Unreal product velocity from the team. https://t.co/OWkksdt10W与 @cerebras 的 @andrewdfeldman 进行的关于快速推理、AI 芯片和下一个计算瓶颈的参考对话也可在 Spotify、Apple 播客和 YouTube 上观看： https://t.co/iP21ZRHgpB
This reference conversation on fast inference, AI chips, and the next compute bottleneck with @andrewdfeldman of @cerebras is also available on Spotify, Apple Podcasts and here on YouTube: https://t.co/iP21ZRHgpB我与@cerebras 首席执行官@andrewdfeldman 的对话。 我们从“什么是晶圆？”开始。并解释了为什么整个芯片行业正在围绕推理速度进行重组。 00:00 冷开场&介绍 01:31 为什么速度成为AI瓶颈 02:32 每个用户每秒的令牌数，解释 03:16 人工智能的宽带时刻和 Netflix 的类比 04:35 人工智能芯片格局：GPU、TPU、Trainium、ASIC 06:36 什么是ASIC？ 08:08 Nvidia、Groq 和快速推理大战 09:16 OpenAI、Broadcom 和专用芯片 12:10 中国、电力和主权人工智能基础设施 15:05 AI基础设施热潮是泡沫吗？ 18:56 隐藏的瓶颈：HBM、CoWoS 和 3nm 22:57 为什么代理商正在创造CPU需求 25:36 Andrew 从 SeaMicro 到 Cerebras 的路径 26:13 为什么 Cerebras 在 2016 年押注人工智能 31:14 SRAM 与 HBM：为什么推理是一个内存问题 33:19 晶圆级计算实际上意味着什么 34:28 深科技“珠穆朗玛峰”问题 36:07 第一个 Cerebras 系统工作的那一刻 36:49 敲响警钟并幸存的深科技 39:08 巨型芯片如何处理故障 41:22 为什么 GPU 难以解码 42:17 预填充与解码的解释 44:01 人工智能推理中的“100部高清电影”问题 45:04 推理如何快速改变强化学习和训练 48:08 推理模型以及为什么它们需要更多计算 50:08 验证、护栏和小模型检查大模型 52:37 多模式人工智能和视频之路 53:51 Cerebras 的商业模式：硬件、云、API 55:14 OpenAI 的 750MW 推理交易 55:36 为什么数据中心以兆瓦来衡量 58:01 AWS Trainium + Cerebras 解码 59:29 快速代币作为云产品 01:00:52 CUDA 仍然是护城河吗？ 01:03:53 台积电如何帮助 Cerebras 打造巨型芯片 01:07:41 为什么2020年没人关心 01:08:15 为什么芯片供应链难以多元化 01:09:54 为什么今天的人工智能模型将是你用过的最糟糕的 01:10:38 快速人工智能可以为 SaaS 带来什么
My conversation with @andrewdfeldman, CEO of @cerebras. We started from "what is a wafer?" and built up to why the entire chip industry is reorganizing around inference speed. 00:00 Cold open & Intro 01:31 Why speed became the AI bottleneck 02:32 Tokens per second per user, explained 03:16 AI’s broadband moment and the Netflix analogy 04:35 The AI chip landscape: GPUs, TPUs, Trainium, ASICs 06:36 What is an ASIC? 08:08 Nvidia, Groq, and the fast inference war 09:16 OpenAI, Broadcom, and specialized silicon 12:10 China, power, and sovereign AI infrastructure 15:05 Is the AI infrastructure boom a bubble? 18:56 The hidden bottlenecks: HBM, CoWoS, and 3nm 22:57 Why agents are creating CPU demand 25:36 Andrew’s path from SeaMicro to Cerebras 26:13 Why Cerebras bet on AI in 2016 31:14 SRAM vs. HBM: why inference is a memory problem 33:19 What wafer-scale computing actually means 34:28 The deep-tech “Everest” problem 36:07 The moment the first Cerebras system worked 36:49 Ringing the bell and surviving deep tech 39:08 How a giant chip handles failure 41:22 Why GPUs struggle with decode 42:17 Prefill vs. decode explained 44:01 The “100 HD movies” problem in AI inference 45:04 How fast inference changes RL and training 48:08 Reasoning models and why they cost more compute 50:08 Verification, guardrails, and small models checking big models 52:37 Multimodal AI and the path to video 53:51 Cerebras’ business model: hardware, cloud, API 55:14 OpenAI’s 750MW inference deal 55:36 Why data centers are measured in megawatts 58:01 AWS Trainium + Cerebras decode 59:29 Fast tokens as a cloud product 01:00:52 Is CUDA still a moat? 01:03:53 How TSMC helped Cerebras build the giant chip 01:07:41 Why nobody cared in 2020 01:08:15 Why chip supply chains are hard to diversify 01:09:54 Why today’s AI models will be the worst you ever use 01:10:38 What fast AI could do to SaaS我们也看到了这一点，并添加了直接使用 claude cli 的代码路径 - 很难与系统对抗。 https://t.co/VM49zPSyk2
We see that as well and added code paths that use the claude cli directly - hard to fight the system. https://t.co/VM49zPSyk2GPT Sol 事件发生后，与一位在一家上市公司负责安全事务的朋友聊天。一些有趣的要点.. 当身份和访问管理是为有限数量的员工设计时，如何有效地管理无限代理的安全性？ 一名员工现在可以启动数百个代理。这些特工可以产生更多的特工。 传统上，每个员工都有身份、角色、权限和生命周期。 代理会继承生成员工的权限吗？他们的生命周期是怎样的——一项任务，一张票，还是一周？子代理是否继承相同的权限？您如何审核这一切？
Chatted with a friend who leads security at a public company following the GPT Sol incident. some interesting takeaways.. How do you manage security for effectively infinite agents when identity and access management was designed for a finite number of employees? One employee can now spin up hundreds of agents. Those agents can spawn more agents. Traditionally, each employee has an identity, a role, permissions, and a lifecycle. Do agents inherit the spawning employee’s permissions? What’s their lifecycle - a task, a ticket, a week? Do child agents inherit the same permissions? How do you audit all of this?鉴于我们如此随意地使用这个标题，科技领域的事物已经失去了所有信号。 > “neo”-某物 > 全栈 > 研究员* > 实验室 > 合作伙伴* > 前沿部署 > RL（慢慢到达那里） * 是的，我知道这很讽刺，因为 a) 我们经营一个奖学金 b) 我自己的头衔是合伙人
Things in tech that have lost all signal given how liberally we use this title.. > “neo”-something > full stack > fellows* > labs > partner* > forward deployed > RL (getting there slowly) * yes I know the irony since a) we run a fellowship and b) my own title is partner