AI
NousResearch Accelerated Business Hackathon
Andrew Ng just dropped a 3-hour course on how to become an AI Engineer in 2026
Sutskever 30 implementations Comprehensive toy implementations of the 30 foundational papers recommended by Ilya Sutskever
用 LLM + 语义聚类，把海量用户评论提炼成四级 VOC 标签体系
DSLs Enable Reliable Use of LLMs
Programming
GPT 5.6 Has 72 Possible Configurations. What’s A Good Default? GPT 5.6 中由不同模型规模与推理期算力组合而成的 72 种配置，并分析了如何基于训练期与推理期的缩放法则（Scaling Laws）在其中寻找最佳的性能与成本权衡
What building Shippy taught us about building agents
让 Agent拥有「跨终端长期记忆」——基于 Amazon Bedrock AgentCore Memory 的实践
Reverse Engineering ChatGPT Web: How OpenAI Built for a Billion Users
OpenWiki 0.2 is adopting the OKF support
weekly AI news digest with GitHub Agentic Workflows
ReactBench v1 an evaluation for coding agents on realistic React work
Agentic Testing: Where Agents Fit in the E2E Testing Stack 传统端到端测试强制执行固定的 UI 路径，而基于 AI 代理（Agentic）的测试则通过动态探索来验证目标。尽管代理测试在执行成本、性能和复杂场景下的稳定性方面存在权衡（Trade-offs），但它们作为更高层级的探索与调试工具，能够有效补充而非替代现有的确定性 CI 测试流水线。
Memex 本地优先、具备类型系统，服务于人与 Agent 的知识工作空间
sail - Drop-in Apache Spark replacement written in Rust
拨开迷雾：炒得火热的FDE究竟是什么？九问九答剖析其是否自带银弹？
Open-source Reflect rewrite A local-first AI agent-friendly Markdown note-taking app
Five studies changing how I think about AI in software engineering