AI with Alex Hexo https://bodysuperman.github.io/ All rights reserved 2026, AI with Alex Attention(Q,K,V)=softmax(QK^T/sqrt(d_k))V AI with Alex's Blog 2026-09-27T02:00:00.000Z AI with Alex https://bodysuperman.github.io/2026/09/06/LangGraph-From-Zero-to-Graph/ 2026-09-06T02:00:00.000Z
  • Introduction
  • Environment Setup
  • What Is LangGraph
  • Core Concepts: The Three Elements and the Runtime Model
  • Graph API vs Functional API
  • Your First Graph: Define, Compile, Invoke
  • State Management
  • Control Flow
  • Edges: A Systematic Summary
  • Node Execution and Fault Tolerance
  • Best Practices
  • FAQ
  • Summary
    • ]]>     LangGraph from Zero: Environment Setup, State Management, and Control Flow in Practice 2026-09-27T02:00:00.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/09/06/LangGraph-From-Zero-to-Graph/ 2026-09-06T02:00:00.000Z
  • 引言
  • 环境准备
  • LangGraph 是什么
  • 核心概念：三要素与运行机制
  • Graph API vs Functional API
  • 第一个图：定义、编译、调用
  • 状态管理
  • 控制流
  • 边的系统总结
  • 节点执行与容错机制
  • 最佳实践
  • 常见问题
  • 总结
    • ]]>     LangGraph 从零开始：环境搭建、状态管理与控制流实战 2026-09-27T02:00:00.000Z     AI with Alex https://bodysuperman.github.io/2026/08/30/Build-your-own-AI-Skills/ 2026-08-30T01:32:54.000Z
  • Introduction
  • What Is a Skill
  • Why Skills Matter
  • Anatomy of a Skill
  • Build Your Own: From Simple to Complete
  • Find Ready-Made Skills
  • Integrating Skills Into Your Workflow
  • Iterating and Versioning
  • Best Practices
  • FAQ
  • Summary
    • ]]>     Build Your Own AI Skills 2026-09-25T16:00:00.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/08/30/Build-your-own-AI-Skills/ 2026-08-30T01:32:54.000Z
  • 引言
  • 什么是技能（Skill）
  • 为什么技能很重要
  • 技能的解剖结构
  • 动手搭建：从简单到完整
  • 寻找现成的技能
  • 将技能融入你的工作流
  • 迭代与版本管理
  • 最佳实践
  • 常见问题
  • 总结
    • ]]>     搭建属于你自己的AI技能 2026-09-25T16:00:00.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/08/23/Deep%E2%80%91Dive-Advanced-Tips-for-Claude-Code/ 2026-08-23T04:48:28.000Z
  • 引言
  • 全局视角：七层扩展体系
  • 选择合适的模型
  • 核心配置
  • 日常命令与交互
  • 正确的工作流：探索、规划、实现、提交
  • 最佳实践
  • 新项目启动套件
  • 常见问题
  • 总结
    • ]]>     深入理解与进阶技巧：Claude Code 2026-09-01T16:00:00.000Z     AI with Alex https://bodysuperman.github.io/2026/08/23/Deep%E2%80%91Dive-Advanced-Tips-for-Claude-Code/ 2026-08-23T04:48:28.000Z
  • Introduction
  • The Big Picture: Seven Layers of Extension
  • Choosing the Right Model
  • Core Configuration
  • Daily Commands and Interactions
  • The Right Workflow: Explore, Plan, Implement, Commit
  • Best Practices
  • Starter Kit for New Projects
  • FAQ
  • Summary
    • ]]>     <![CDATA[Deep‑Dive & Advanced Tips for Claude Code]]> 2026-09-01T16:00:00.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/08/06/Six-money-saving-tips-for-Claude-Code/ 2026-08-06T08:15:59.000Z Anthropic 发布了一篇博客，分享了六个节省 Claude Code 开支的技巧：完成任务后用 /clear 清除对话；在开始时锁定模型和推理强度，避免中途切换导致提示缓存失效；用 @ 引用文件而非手动输入路径；为产生大量输出的命令添加静默参数；离开前运行 /compact；将大输出量的任务分配给子代理。

    pic

    ]]>     六个节省 Claude Code 开支的技巧 2026-08-16T10:04:19.000Z     AI with Alex https://bodysuperman.github.io/2026/08/06/Six-money-saving-tips-for-Claude-Code/ 2026-08-06T08:15:59.000Z Anthropic released a blog sharing six money-saving tips for Claude Code: clear the conversation with /clear after completing a task; lock the model and reasoning intensity at the start to avoid invalidating the prompt cache by switching; use @ to reference files instead of typing paths manually; add silent parameters to commands that generate a lot of output; run /compact before taking a break; assign large output tasks to sub-agents.

    pic

    ]]>     Six money-saving tips for Claude Code 2026-08-16T10:04:19.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/06/25/From-Post-Training-to-Real-Time-Perception-Self-Evolving-Agents-and-Multimodal-Interaction/ 2026-06-25T07:09:38.000Z
  • 引言
  • 后训练：将能力烘焙进参数
  • 自进化智能体：从经验中学习
  • 实时感知与多模态交互
  • 整合：现代智能体的全栈
  • 最佳实践
  • 常见问题
  • 总结
    • ]]>     从后训练到实时感知：自进化智能体与多模态交互 2026-07-25T07:09:38.000Z     AI with Alex https://bodysuperman.github.io/2026/06/25/From-Post-Training-to-Real-Time-Perception-Self-Evolving-Agents-and-Multimodal-Interaction/ 2026-06-25T07:09:38.000Z
  • Introduction
  • Post-Training: Baking Capability Into Parameters
  • Self-Evolving Agents: Learning From Experience
  • Real-Time Perception and Multimodal Interaction
  • Putting It Together: The Full Stack of a Modern Agent
  • Best Practices
  • FAQ
  • Summary
    • ]]>     From Post-Training to Real-Time Perception: Self-Evolving Agents and Multimodal Interaction 2026-07-25T07:09:38.000Z     AI with Alex https://bodysuperman.github.io/2026/06/17/From-Capability-to-Intelligence-Tools-Coding-and-Evaluation-in-AI-Agents/ 2026-06-17T07:09:38.000Z
  • Introduction
  • What: The Three-Layer Model of Agent Intelligence
  • Tools: The Agent’s Hands, Feet, and Senses
  • Coding Agent: Code as the Meta-Capability
  • Evaluation: Knowing Whether It Actually Works
  • Example: An End-to-End Mini-Case
  • Best Practices
  • FAQ
  • Summary
  • Appendix
    • ]]>     From Capability to Intelligence: Tools, Coding, and Evaluation in AI Agents 2026-07-24T07:09:38.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/06/17/From-Capability-to-Intelligence-Tools-Coding-and-Evaluation-in-AI-Agents/ 2026-06-17T07:09:38.000Z
  • 引言
  • 是什么：Agent 智能的三层模型
  • 工具：Agent 的手脚与感官
  • 编程 Agent：代码作为元能力
  • 评估：如何知道它到底行不行
  • 示例：一个端到端的迷你案例
  • 最佳实践
  • 常见问题
  • 总结
  • 附录
    • ]]>     从能力到智能：AI Agent 的工具、编程与评估 2026-07-24T07:09:38.000Z     AI with Alex https://bodysuperman.github.io/2026/06/10/Building-AI-Agents-From-Context-to-Memory/ 2026-06-10T07:09:38.000Z
  • Introduction
  • What Is an AI Agent
  • Context Engineering: The Ceiling of Agent Capability
  • From Context to Memory: Making Agents Remember
  • Harness Engineering: Reliability Beyond the Demo
  • Best Practices
  • FAQ
  • Summary
    • ]]>     Building AI Agents: From Context to Memory 2026-07-23T07:09:38.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/06/10/Building-AI-Agents-From-Context-to-Memory/ 2026-06-10T07:09:38.000Z
  • 引言
  • 什么是 AI Agent
  • 上下文工程：Agent 能力的天花板
  • 从上下文到记忆：让 Agent 拥有记忆力
  • Harness 工程：超越 Demo 的可靠性
  • 最佳实践
  • 常见问题
  • 总结
    • ]]>     构建 AI Agent：从上下文到记忆 2026-07-23T07:09:38.000Z     AI with Alex https://bodysuperman.github.io/2026/05/18/LangChain-Runtime-Middleware-Core-Principles-and-more/ 2026-05-18T04:00:00.000Z
  • Introduction
  • What
  • Why
  • How
    •  <![CDATA[LangChain Runtime & Middleware: Core Principles and Practical Guide]]> 2026-06-03T04:00:00.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/05/18/LangChain-Runtime-Middleware-Core-Principles-and-more/ 2026-05-18T04:00:00.000Z
  • 引言
  • 是什么
  • 为什么
  • 怎么做
    •  LangChain 运行时与中间件：核心原则与实践指南 2026-06-03T04:00:00.000Z     AI with Alex https://bodysuperman.github.io/2026/05/11/A-Multimodal-AI-Agent-Application-AI-Private-Chef/ 2026-05-11T04:48:28.000Z
  • Introduction
  • What Are We Building?
  • Implementation Strategy
  • Prototyping in Jupyter
  • Deploying with LangGraph and LangSmith
    •  A Multimodal AI Agent Application: AI Private Chef 2026-06-09T16:00:00.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/05/11/A-Multimodal-AI-Agent-Application-AI-Private-Chef/ 2026-05-11T04:48:28.000Z
  • 引言
  • 我们要构建什么？
  • 实现策略
  • 在 Jupyter 中制作原型
  • 使用 LangGraph 与 LangSmith 部署
    •  多模态 AI Agent 应用实战：AI 私厨 2026-06-09T16:00:00.000Z     AI with Alex https://bodysuperman.github.io/2026/05/04/LangChain-Core-Components%EF%BC%9AMode-Message-Prompt-Tool-Memory/ 2026-05-04T04:48:28.000Z
  • What is LangChain?
  • Models — The Brain
  • Messages — The Building Blocks of Conversation
  • Prompts — Steering the Output
  • Tools — The Hands
  • Memory — Making the Agent Remember
    •  LangChain Core Components: Model, Message, Prompt, Tool, Memory 2026-06-05T16:00:00.000Z     AI with Alex https://bodysuperman.github.io/zh-CN/2026/05/04/LangChain-Core-Components%EF%BC%9AMode-Message-Prompt-Tool-Memory/ 2026-05-04T04:48:28.000Z
  • 什么是 LangChain？
  • 模型（Model）—— 大脑
  • 消息（Message）—— 对话的基本单元
  • 提示词（Prompt）—— 引导输出
  • 工具（Tool）—— 双手
  • 记忆（Memory）—— 让 Agent 记住一切
    •  LangChain 核心组件：模型、消息、提示词、工具、记忆 2026-06-05T16:00:00.000Z