Latest

Every published item, in order · 5281 items
Oct 22
Tomorrow 08:00, the daily lands in your inboxTelegram channel or email. Same screening as this page, no ads.
Chinese content实测 72 小时抓包:手机 App 偷听?麦克风调用次数为 0,真凶另有其人

作者真拿 iPhone 15 Pro 和 Pixel 8 抓包 72 小时,41.8GB 流量跑下来麦克风唤醒次数是 0,然后把 PSI、端侧 Embedding 这些真正干活的追踪技术一层层拆给你看,比讲 Cookie 的老科普硬核多了。

掘金本周最热@第404号工位Source
How-toBuildersOfficial05:47
Chinese content用 Strands Agents 构建自适应智能体框架:本地 SLM 处理简单任务,复杂请求走 Amazon Bedrock

AWS 教程教你怎么用 Strands Agents 搭一套混合架构:简单活儿让 Ollama 本地小模型干,难活儿丢给 Bedrock 云端模型,还带 IoT 监控实例。

Event
Primary sourceAWS Machine Learning Blog@Nitin EusebiusSource
How-toBuildersOfficialSelected04:08
AWS shows three ways to train Qwen3 on Trainium with torchtitan

The notable point is that one config setup now covers three different training backends on Trainium, so teams can switch between them without rewriting launch scripts or breaking checkpoint compatibility. Engineers choosing between NVIDIA GPUs and AWS silicon can use this to evaluate Trainium for fine-tuning open-source models like Qwen3. The guide also shows how NKI MoE kernels fit into the same workflow.

Event
Primary sourceAWS Machine Learning Blog@Josh LongeneckerSource
How-toBuildersSelected03:43
PyTorch Examines Linear Algebra Kernels Behind Faster Optimizers

The speed of an optimizer depends as much on kernel-level linear algebra as on step count, a factor most training guides ignore. Engineers tuning large-scale training can apply the QR, Cholesky and eigendecomposition implementation patterns to cut data movement on GPUs. The talk ties directly to GPU MODE's competition, giving concrete reference implementations to study.

Official accountPyTorch@PyTorchSource
How-toBuildersSelected02:26
Kena's web index cuts agent search output tokens by 50%

Token-level scoring over its own index means agents fetch only the passages that matter, which lowers API costs for teams running frequent web searches. Builders wiring up agents through MCP or API can swap in KeenableAI as a search backend and immediately halve the tokens their pipeline pays for. Compared with generic search APIs that return whole snippets, this targets long-document retrieval where most waste happens.

Ate-a-Pi@svpinoSource
How-toBuildersMulti-sourceSelected01:01
Anyscale session covers Ray Serve LLM deployment and inference scaling

The session bundles inference infrastructure decisions most teams learn the hard way: when disaggregated prefill/decode and KV-cache-aware routing pay off, and whether self-hosting beats token-based API pricing. Engineers deploying open-source models with vLLM or SGLang can register to get a hands-on walkthrough of Ray Serve LLM and agentic-platform integration. Registration is open now via the Anyscale link.

Event
Anyscale@anyscalecomputeSource
Oct 21
How-toBuildersSelected18:37
dsh-context Tool Visualises Agent Context Usage for DSH

Context cost in long-running agents usually hides inside tool results and sub-agent calls, and dsh-context makes those allocations traceable instead of guessing from a total token count. Builders running multi-agent workflows on DSH can use it to diagnose cache misses, runaway context growth and per-sub-agent spend, then rebalance prompts or trimming before costs climb.

AIGCLINK@aigclinkSource

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