Skip to content

Dev weekly 2026-Sep-07

Published: at 02:02 AM

AI and Programming

TimesFM-3: A zero-shot foundation model for multivariate forecasting

CS 329Z: Engineering AI Agents

zg 正式开源:本地检索,不止于关键词 https://zvec.org/en/docs/zvec-grep/embedding-models/

AI Search - free lesson

AI 搜尋前沿技術總覽: 哪些該做、哪些值得投入、哪些先知道就好

Claude 商务智能体

智能体请求暴增 9.4 倍,token 账单却没涨:Uber 公开 AI 软件工厂省钱方法

Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI

OpenScholarXIV

Stop Letting Agents Run the Workflow

WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution

FrontierHarness Eval

zero-to-agent-30-Oreilly-FinanceAgent

AI Engineer Notebooks

How Warp builds self-improving agents on Claude

Anthropic最新多智能体研究 观察到了三类系统性的失效模式 Patterns and problems in emerging multiagent systems https://www.anthropic.com/research/multiagent-systems

How people ask Claude for personal guidance

Agent-Native is an open-source TypeScript framework for building autonomous agents with intuitive UIs

Tangle - Build ML and data pipelines collaboratively

One Skills Repo, Every AI Agent: How Our Data Team Ships Its Conventions to Claude Code, GitLab Duo, OpenCode, and Snowflake CoCo

Node.js Interactive 2026: A Recap

shadcn 通过新推出的聊天组件将对话基础组件引入了 shadcn/ui

如何使用langsmith对ai agent和prompt进行测试(类似unit test或者integration test) “Dataset → Target → Evaluators → Experiment → Regression Comparison”,所以测试的是:output 是否满足 expectations

Snowflake Postgres: Zero-ETL from OLTP to Analytics

Others

阿里投资Kimi始末 - elsewhere

duck.ai by duckduckgo

人之恶,在好为人师

智谱与MiniMax背对背拥抱

Gartner Survey Finds Only 22% of Organizations Have Successfully Scaled AI Across Multiple Business Units