开启时会模糊预览图，关闭后正常显示
anthropic blog: https://t.co/BdPHMReRL3
building workers observability @cloudflaredev, prev founder @baselimehq (acquired by cloudflare), prev aerodynamicist
the anthropic blog focuses on how this could be used within a filesystem and paired with mcp and sandboxed code excecution I like to keep things much simpler, this runs in a @cloudflaredev durable object, I'm yet to try the new dynamic worker loader
anthropic blog: https://t.co/BdPHMReRL3
anthropic just published a blog highlighting a pattern I've been using for weeks: a search tools tool instead of loading all tools into the context window of the model, I load a "search tools" tool the model can use to browse a catalog of tools just-in-time tool discovery
the anthropic blog focuses on how this could be used within a filesystem and paired with mcp and sandboxed code excecution I like to keep things much simpler, this runs in a @cloudflaredev durable object, I'm yet to try the new dynamic worker loader
这个内容还是挺有意思的 他在想办法论证评估AI做为一种元能力的渗透占比 当这个渗透占比攀高时，将进入指数增长期 可能以前的很多事情都会快速失效，AI会产生巨大影响
Believing is seeing
在学习完生成式 AI 技术的相关理论基础后，如何应用到实际项目当中，让我们颇为烦恼。 恰巧，发现了 Google 推出的 “面向开发者的生成式 AI” 学习进阶路线，提供了丰富的动手实操机会，全面了解从模型开发到生产应用的完整流程。 不仅包括生成式 AI 的核心技术栈，从 Transformer 架构、注意力机制到图像生成模型，还包括负责任 AI 的实践方法和 MLOps 部署技巧。 课程地址：https://t.co/pCO1OsYzcg 主要内容： - 图像生成和扩散模型：学习如何使用扩散模型生成高质量图像； - Transformer 核心架构：深入理解注意力机制、编码器-解码器架构和 BERT 模型； - Vertex AI Studio 实践：掌握 Google 的生成式 AI 工具，从原型到生产； - 向量搜索与 RAG：学习语义搜索和检索增强生成技术，减少 AI 幻觉； - 负责任 AI 实践：涵盖公平性、可解释性、隐私和安全等关键主题； - MLOps 部署：了解生成式 AI 模型的运维和管理最佳实践。 课程面向有一定基础的开发者进阶学习，初学者建议先去学习 “生成式 AI 简介” 这门课程。
💡 挖掘开源的价值 🧑🏻💻 坚持分享 GitHub 上高质量、有趣、实用的教程、AI工具、前沿 AI 技术 🧐 A list cool, interesting projects of GitHub. ✏️ 公众号：GitHubDaily
> Their solution to this problem is to mitigate their own risk by demanding the rest of the EU agree to helping them repay the Russians in the event all of this goes south I don't understand the issue tbh. The solution is clearly to recognize that Orcs have no legal rights.
The recipe of the just future has already been provided by Krylov in 1997.