跳到正文
今天10月24日周六14 条
  1. meng shao71

    作者解读 Shopify Engineering 文章:Shopify 为 Slack 里的 AI Agent「River」搭建 Remote Dev,用远程 Mac 机群加挂载 Linux 的 Android 模拟器,让 Agent 并行编译、测试 iOS/Android、截图录屏并做 Live Preview。

    引用Mustafa Ali@mustafa01ali

    Local machines had become a bottleneck for agentic mobile development at @Shopify, so we gave agents their own remote dev environments. They now author 61% of our merged PRs 🧵

  2. Databricks:Blog50

    Databricks 实测:Spark 声明式管道让 AI 智能体写数据管道更省更安全

    Databricks 用 Opus 4.8 在 22 个数据工程任务、264 次构建上对比 Apache Spark 的过程式与声明式管道,声明式在运行前即可校验结构错误,执行器耗时从 596 秒降至 17 秒,约省 34 倍算力。声明式管道代码量中位数为 64 行,约为过程式 131 行的一半,且写法更统一。两者完成率接近,分别为 258/264 和 255/264,逻辑错误表现持平。

  3. OpenAI:官网动态54

    OpenAI 详解 Habitat 存储平台如何支撑超 10 亿 ChatGPT 用户

    OpenAI 发布 Habitat 存储平台系列第二篇,讲解其为服务超 10 亿 ChatGPT 用户如何扩展底层 Azure Cosmos DB。平台在 40 个地理区域托管超 500 PB 数据,经历了从共享多租户到单租户容器再回到多租户默认的循环,与 Azure 团队通过热键限流、QoS 分级和新版 serverless 模式解决噪音邻居和弱分区问题。

  4. meng shao60

    作者转述 Grok Bot 团队 Matt Palmer 对 Team Bots 的讲解:Team Bot 即分享给团队的 Grok Bot,把个人流程封装成同事可调用的内部工具,支持权限分层、接入 Slack/Teams,聊天互相私密。

    引用matt palmer@mattyp

    Team Bots - Grok Bots for your whole team 1:18 A simple example 1:51 Why they're powerful 2:32 Connecting your tools 3:11 How to make one 3:52 A complex example 5:48 When to make one 7:21 Under the hood 7:55 Recap Currently available to Teams & Enterprise customers - more soon!

10月23日周五
  1. 公众号:小红书技术49

    小红书推荐系统评测 Agent:上下文驱动的 LLM-as-Judge 实践

    小红书构建了上下文驱动的推荐评测 Agent,以一次推荐请求为最小评测单元,通过确定性规则与 LLM-as-Judge 分工,对用户当时看到的内容做分维度语义判断。系统采用 0/1/2 三档评分,并固定输入快照与评测版本,使结果可比较、可回放、可下钻到具体请求与单篇内容。团队还探索 LLM-as-User 连续用户模拟,目前定位为上线前的离线预筛工具,不替代线上 A/B 实验。

  2. meng shao47

    SkyRL 通过四类系统优化,在 B200(约 178 GiB)上把 GLM-5.3-Flash 的单序列训练长度做到 100 万 token,峰值约 144.5 GiB,约为开箱 Megatron 基线的 64 倍。

    引用avi basnet@avigyabasnet

    Super excited to share some of the work I’ve been doing on supporting 1M token context training on SkyRL for GLM 5.3 Flash, 64x higher than the out of the box Megatron baseline! 🚀 This work was done in collaboration with the amazing folks at @LilaSciences! 🤩 Getting there took key algorithmic and systems optimizations in four main areas: 1) DeepSeek Sparse Attention: run DSA on a sparse kernel; chunk and shard its quadratic indexer. 2) Activation recompute and offloading: recompute activations in backward and offload saved inputs to CPU. 3) Context parallelism: split each sequence across GPUs, with new CP support for KDA and a fix for DSA. 4) Avoiding unnecessary precision upcasts: keep FP32 copies out of memory and off the network. For more details on our implementations and our results, check out our full technical blog here: https://blog.skyrl.ai/blog/1m-context-glm-5-3-flash! This work benefited from the contributions of many people along the way. Big thanks to @pcmoritz, @erictang000, @sumanthrh, @CyrusHakha, @richliaw from @anyscalecompute and Keith Truongcao, Tushar Khot, Vivek Subramanian, Semih Kara from Lila Sciences for all the help and guidance!

  3. Figma AI / Agent / Make51

    Figma 官方讲解如何让 Figma agent 用好你的设计系统

    Figma 官方介绍已全面开放的 Figma agent 如何借助设计系统产出符合团队规范的设计。方法包括连接已发布的组件库、上传 guidelines 固化团队规则、用 Examples 页面或 _example 命名提供实际范例、把重复流程做成 skills(如 /design-system-audit 和 /analyze-components)。

  4. lauren23

    最近有点安静——我在做一个新的配套技能,能帮你的智能体画出这些漂亮的状态图/图表,并用你在 pstack 里通过 /create-formal-verification-skill 做好的规范来校验它们! 它现在还用了 @kepano 的 Pele,所以看起来才这么好看。 很快发布!

    引用lauren@poteto

    there's something really cool about asking your agent to draw a statechart (hypothetical example in this image), then using the /create-formal-verification-skill you made with pstack to find bugs through something like TLA+. you can both understand your code better and how the user flow should work, and fix issues users would otherwise see