Latest

Every published item, in order · 55860 items
Nov 12
How-toBuildersSelected04:00
PyTorch Maintainers Show Single-GPU Testing for Distributed Training

The non-obvious point is that most distributed-training failures can be caught on a single GPU with stubbed communication and static SPMD checks, saving cluster hours. Engineers preparing large PyTorch runs, or teams with limited GPU access, can apply the fake process group pattern directly. The same keynote also introduces torch_remat and FlexShard, worth tracking for memory-efficiency workloads.

Official accountPyTorch@PyTorchSource
Chinese contentMarc Andreessen谈AI编程:顶尖AI程序员年收入达5000万美元

Andreessen 在 Joe Rogan 节目里聊的这几点很扎心:AI 编程没让人下班更早,反而催生了通宵干活的“AI vampire”。后面关于用决策模型自动路由 prompt 的思路,对每天开好几个标签页调模型的开发者很实用。

rohanpaul_ai@rohanpaul_aiSource
Chinese contentCodeRabbit 上架 Antigravity 插件市场,代码审查与智能体编写同流程完成

CodeRabbit 官宣进 Antigravity 插件市场,智能体写完代码它就在同一流程里跑 review,不用切工具。之前审查和生成是两套系统,现在合在一起,用 Antigravity 的团队装个插件就行。写代码靠智能体的朋友可以试试。

coderabbitai@coderabbitaiSource
IndustryUS vs ChinaMulti-source03:42
Chinese contentRamp AI Index:10月美国企业AI支出增速放缓至1.65%

Ramp 追踪的美国企业AI支出10月只涨了1.65%,增速明显放缓。更扎眼的是1%的公司撑起 OpenAI 和 Anthropic 八成收入,说明重度用户之外渗透并不顺利。做SaaS、投AI的人建议看看这组数据,判断客户支持场景能不能接棒编程业务。

Event
Official accountDecoder@Matthias BastianSource
How-toBuildersMulti-source03:38
Chinese contentSimon Willison 为 244 个 vibe-coded 工具站新增标签浏览功能

Simon Willison 把自己攒的 244 个 vibe-coded 小工具加上了标签云,靠 Fable 5.2 加 tree-sitter 自动读代码提取标签。想批量整理自己乱堆的小项目或做个工具站的人可以直接抄这个工作流。

Event
Official accountSimon Willison’s WeblogSource
PapersBuildersSelected03:31
Perplexity red-teams SPACE sandbox with 12 AI models

The counterintuitive result — general-purpose models beating cyber-specialized ones at sandbox escapes — suggests teams should benchmark attackers on their own infrastructure rather than trusting security-tuned model variants. Builders of computer-use agents and sandbox environments can apply the prompting guidance and DNS spoofing finding directly when hardening isolation boundaries. Open-source access to SPACE makes it a practical starting point for running comparable red-team evaluations.

Aravind Srinivas@AravSrinivasSource
Tomorrow 08:00, the daily lands in your inboxTelegram channel or email. Same screening as this page, no ads.
IndustryUS vs ChinaMulti-source03:25
Chinese contentNvidia 黄仁勋相关指控再引中美 AI 芯片管制讨论

CSET 把两件事放在一起讲:黄仁勋相关指控背后的中国情报疑虑,加上美国本地反对数据中心的风潮,后者可能拖慢美国算力扩张。做芯片、算力或政策研究的读者可以看看它如何主张用证据而非情绪来定政策。

Event
CSET Georgetown@CSETGeorgetownSource
Chinese contentDeloitte 解析 Google 全栈垂直整合的 AI 优势

如果你们公司在选云上 AI 方案,这篇借 Deloitte 专家的口把 Google 和只卖模型的对手比了比:从自研 TPU 芯片一路整合到 Gemini 应用,性能和单位成本都有优势。做采购或技术选型的人可以拿来当参考视角,不过注意它是 GeminiApp 赞助内容。

The Information@theinformationSource
Chinese contentAWS CEO Matt Garman:太空数据中心至少还需 10 到 20 年

AWS 的掌门人直接给太空数据中心泼了盆冷水:全球连一天一发的火箭都没有,一机架 GPU 就有一吨重,太空中散热和维修更是硬伤。他把账算得很细,做算力基础设施判断、评估这类创业项目时很值得对照着看。

rohanpaul_ai@rohanpaul_aiSource
IndustryUS vs ChinaMulti-source03:12
Chinese contentRamp 数据显示 1% 企业贡献 OpenAI 和 Anthropic 80% 收入

Ramp 经济学家用自家企业消费数据拆了 OpenAI 和 Anthropic 的收入结构:1% 客户贡献八成收入。中位数企业每月每员工只花 12 美元,头部却花 1 万美元,差距近千倍。想看清 AI 商业化真实渗透率、判断该不该给团队配订阅的,都值得看看这些分层数字。

Event
rohanpaul_ai@rohanpaul_aiSource
Chinese contentBlue 发布可对话操控的 AI 手机,智能体可后台使用所有应用

编码智能体有专属环境后,终于有人把这套思路搬到手机上。Blue 让智能体跑在自己的 OS 里,能操作全部应用、分拣通知,说一句话就帮你交报销单。团队称 10 位 iPhone 用户换机后没人想回去。经常被消息和待办淹没的人可以关注。

elvis@omarsar0Source
ProductsBuildersSelected03:08
ElevenLabs adds Routines to ElevenAgents for follow-up tasks

Voice agents that can execute pre- and post-call steps shift the build burden from custom orchestration code to platform configuration, shortening deployment timelines for support and sales teams. Teams evaluating ElevenAgents against alternatives like Vapi or Retell should compare how each handles callbacks and SMS in native workflows. Operations leads in fintech and telecom with high call volumes stand to gain the most from automated follow-through.

Official accountElevenLabs@elevenlabsioSource
ProductsBuildersMulti-source02:54
Chinese contentCoreWeave Database 正式发布:托管 PostgreSQL 直接跑在 CKS 集群内

CoreWeave 把托管 PostgreSQL 塞进了自己的 CKS 集群里,基于 CloudNativePG,还带 pgvector,做 Agent 和 RAG 不用再跨云连数据库了。在 CoreWeave 上跑 GPU 工作负载的团队可以省掉一跳网络延迟,跟自己搭数据库或用 AWS RDS 的方案比,数据离计算更近。

Event
CoreWeave@CoreWeaveSource
Chinese contentYC 团队推出语音控制 AI 手机 Blue,跨应用保持上下文

YC 有个团队做了台语音操控的手机 Blue,不是重写系统,而是在你现有应用上跑一个智能体,能跨应用记住上下文,说句“回她周四的事”就能自动找到对话并起草回复。跟苹果 Apple Intelligence 那种单应用内助手思路不一样,适合想抢先看手机交互下一形态的人。

rohanpaul_ai@rohanpaul_aiSource
ProductsBuildersSelected02:43
Databricks makes Genie Ontology generally available

The 1.6× accuracy gain at under half the tokens suggests ontology grounding may matter more to agent reliability than model choice itself. Enterprise teams already running Unity Catalog can layer this on without new governance plumbing, making it a practical test case for agent accuracy work. Power BI and Google Drive connectors put it in direct competition with Microsoft's Fabric knowledge tooling.

Event
Official accountDatabricks@databricksSource
IndustryBuildersSelected02:42
Perplexity fixes network flaws agents exploited in SPACE sandbox

The takeaway is that sandbox hardening was done at the network layer, and the vendor explicitly treats escape testing as ongoing because stronger models may find new routes. Teams building or evaluating agent execution environments should track the full research post for technical details and re-audit their own network policies. The work signals that agent security benchmarking is shifting from one-off audits to continuous testing as model capabilities improve.

Official accountPerplexity@perplexity_aiSource
ProductsBuildersMulti-source02:42
Chinese contentElevenLabs 在 Summit 预告 ElevenAI 工作助理,开放早鸟申请

ElevenLabs 这次把语音能力从客服搬到了办公桌前:任意应用里语音输入、自动会议笔记、通话实时提醒都有。做产品或运营的如果每天开会多,可以申请 early access 试试和 Otter 这类工具的差别。

Event
Official accountElevenLabs@elevenlabsioSource
Chinese content2026 科技就业报告:岗位总量大涨,但增长几乎不在软件岗

Lenny 每半年一期的科技就业报告,这次数据最反常:岗位总量涨 73%,但工程岗原地踏步,UX 研究还砍了 45%。真涨的是数据中心运维和硬件。想搞清楚 AI 时代该往哪个岗位方向走,这份数据比直觉靠谱。

Lenny Rachitsky@lennysanSource
Chinese contentGoogle DeepMind AlphaProof Code 四周完成 Signal 加密核心验证

DeepMind 用 AlphaProof Code 给 Signal 的加密代码做了数学证明,四周搞定过去要几年的活,600 万行证明代码只有 1% 需要人看。做安全或写密码学库的团队应该关注,这套方法以后可能用来验证电网和医疗设备软件。

Official accountDecoder@Manuel UthSource
ModelsUS vs ChinaMulti-source83°02:41
Chinese contentClaude Fable 5.2 (Max) 登顶 Agent Arena,净改进领先第二名 30.4%

Anthropic 这版把 Agent Arena 的分数直接拉开了一倍多差距,Code 和 Work 也都冲到第一,API 还降价 20%。做智能体和编程工作流的朋友可以试试,尤其在意可控性的场景;预算紧的话 GPT-6.1 Terra 每任务不到 1 美元,性价比更合适。

Event
lmarena.ai@lmarena_aiSource
ProductsUS vs ChinaMulti-source02:41
Chinese content字节跳动 Seedance 2.5 4K 版上线 Runway 平台

字节的 Seedance 2.5 现在能在 Runway 里直接生成 4K 视频了,不用再切去即梦。Runway 走多模型路线,和 Veo、Sora 在第三方分发上抢创作者。做视频的朋友可以在一个工作流里混用几个模型,值得试试 4K 档的实际效果。

Event
Official accountRunway ML@runwaymlSource
How-toBuildersSelected02:40
Perplexity Finds CTF Tasks Poorly Measure Agent Security Skills

The non-obvious point is that even security-specialised models can underperform on real red-teaming work, so vendor claims of security fine-tuning deserve scrutiny. Teams evaluating agents for security-sensitive deployments in finance, infrastructure or SaaS should design internal task suites instead of relying on CTF leaderboard scores. The finding applies regardless of region, since evaluation methodology, not model origin, is the bottleneck.

Official accountPerplexity@perplexity_aiSource
PapersBuildersOfficialSelected02:40
Apple Details Weighted State Souping for Long-Context SSM Retrieval

The non-obvious point is that cached SSM states can be reused like precomputed retrieval index entries, so combining them with learned weights changes how memory-heavy applications manage context cost. Builders of agents, RAG pipelines and personalized assistants can evaluate this as a cheaper alternative to re-encoding long contexts on every query. Compared with transformer-based long-context methods, it trades architectural familiarity for fixed memory and faster inference.

Primary sourceApple ML ResearchSource
Chinese contentMAI Transcribe 2 上线 OpenRouter 后请求量增长 12 倍

微软的语音转写模型在 OpenRouter 上单周跑了 850 万次请求,两个月涨了 12 倍,连整个转写类目都跟着涨了 3 倍。做语音功能、在挑转写 API 的开发者可以看看这份真实用量排名,比跑分榜单更接近实际选择。

Event
OpenRouter@OpenRouterAISource
PapersBuildersOfficialSelected02:40
Apple Research Links Gating to In-Context Learning in State Space Models

The non-obvious takeaway is that SSM weaknesses on retrieval may be fixable through gating design rather than requiring a wholesale return to Transformers. Model architects and ML engineers evaluating efficient alternatives to attention should study the paper's analysis to guide architecture choices. Compared with prior empirical comparisons between SSMs and Transformers, this work offers a mechanistic explanation rather than just benchmark observations, which is useful for anyone building long-context inference systems.

Primary sourceApple ML ResearchSource

Only the latest 2000 items are listed; for older ones see the daily archive