🗺️ 课程 004 · 导读与大纲全景（About this course）
官方课程名：Building with the the AI API 官方地址：
https://anthropic-partners.skilljar.com/claude-with-the-anthropic-api针对认证：CCAR-F（架构师 Foundations） — 本库当前主攻 抓取日期：2026-10-15 归档路径：Claudeexams/20-课程/课程004 Building with the the AI API/语料来源：官方 Skilljar 伙伴站公开课程页（About 全文与 Curriculum 无需登录；课时正文需登录）
📄 官方 About（英文原文 ＋ 精译）
Course Overview
官方原句：This comprehensive video course teaches developers how to integrate the AI into applications using the the AI API. The curriculum covers fundamental API operations, advanced prompting techniques, tool integration, and architectural patterns for building AI-powered systems. Through hands-on exercises and practical examples, participants will learn to implement conversational AI, retrieval-augmented generation, automated workflows, and leverage the AI's multimodal capabilities for processing text, images, and documents.
精译：这门综合视频课教开发者如何用 the AI API 把 the AI 集成进应用。课程覆盖基础 API 操作、进阶提示技巧、工具集成，以及构建 AI 系统的架构模式。通过动手练习与实例，学员将学会实现对话式 AI、检索增强生成（RAG）、自动化工作流，并利用 the AI 的多模态能力处理文本、图像与文档。
页头短描述（meta / og:description）
官方原句：This comprehensive course covers the full spectrum of working with the AI provider models using the the AI API
精译：这门综合课覆盖使用 the AI API 调用 the AI provider 模型的全谱系内容。
What You'll Learn（官方 12 条）
| # | 官方原句 | 精译 | 命中 CCAR-F 领域 |
|---|---|---|---|
| 1 | Set up and authenticate with the the AI provider API, including API key management and request configuration | 完成 the AI provider API 的接入与鉴权，含密钥管理与请求配置 | D5 |
| 2 | Implement single and multi-turn conversations with proper message formatting and context handling | 实现单轮与多轮对话，正确处理消息格式与上下文 | D1 / D5 |
| 3 | Configure system prompts and control model behavior using temperature, response streaming, and structured output formats | 配置 system prompt，并用 temperature、流式响应、结构化输出控制模型行为 | D3 |
| 4 | Design and execute prompt evaluation workflows with test dataset generation and automated grading systems | 设计并执行提示评估工作流，含测试集生成与自动打分 | D3 |
| 5 | Apply prompt engineering techniques including XML tag structuring, example-based learning, and clear directive formulation | 运用提示工程技巧：XML 标签结构化、示例学习、清晰指令 | D3 |
| 6 | Integrate the AI's tool use capabilities to extend functionality with custom tools, batch operations, and web search | 集成工具调用能力，用自定义工具、批量操作与网页搜索扩展功能 | D4 |
| 7 | Build retrieval-augmented generation (RAG) systems with text chunking, embeddings, BM25 search, and contextual retrieval | 构建 RAG 系统：文本分块、向量嵌入、BM25 检索、上下文检索 | D1 |
| 8 | Utilize the AI's extended features including extended thinking mode, image analysis, PDF processing, and citation generation | 使用扩展能力：扩展思考、图像分析、PDF 处理、引用生成 | D3 / D5 |
| 9 | Implement prompt caching strategies to optimize API usage and reduce latency | 实施提示缓存策略以优化用量、降低延迟 | D5 |
| 10 | Develop Model Context Protocol (MCP) servers and clients for standardized tool and resource integration | 开发 MCP server 与 client，实现标准化的工具与资源集成 | D4 |
| 11 | Deploy the AI provider Apps including AI Code Assistant for automated development tasks and Computer Use for UI automation | 部署 the AI provider Apps：AI Code Assistant（开发自动化）与 Computer Use（UI 自动化） | D2 |
| 12 | Architect agent-based systems with parallelization, chaining, and routing workflows | 架构基于智能体的系统：并行化、链式、路由工作流 | D1 |
⭐ 判读：12 条里有 5 条直指 D1/D3 的高权重区；第 12 条「Architect agent-based systems」是官方对架构师角色的直接点名。
Prerequisites（官方）
官方原句：Proficiency in Python programming ／ Basic knowledge of handling JSON data
精译：Python 编程熟练；JSON 数据处理基础。
Who This Course Is For（官方 7 类）
官方原句：Backend developers building AI-powered APIs and services ／ Full-stack engineers integrating conversational AI into web applications ／ Data engineers implementing document processing and knowledge retrieval systems ／ DevOps professionals automating workflows with AI assistance ／ Technical architects designing scalable AI-integrated systems ／ Software engineers transitioning to AI/ML application development ／ Developers working on chatbots, virtual assistants, or content generation tools
精译：构建 AI 驱动 API 与服务的后端开发者；把对话式 AI 集成进 Web 应用的全栈工程师；做文档处理与知识检索的数据工程师；用 AI 自动化工作流的 DevOps；设计可扩展 AI 集成系统的技术架构师；转向 AI/ML 应用开发的软件工程师；做聊天机器人、虚拟助理或内容生成工具的开发者。
📚 Curriculum 全景（13 章节 / 85 课时）
图例：🎬 视频 ｜ 📄 图文 ｜ 📝 测验 直链形如
https://anthropic-partners.skilljar.com/claude-with-the-anthropic-api/<lesson-id>；打开正文需 Partner SSO 登录。
| 总序 | 章节 | 类型 | 课时 | 直链 | lesson ID |
|---|---|---|---|---|---|
| 1 | 1 | 📄 图文 | Welcome to the course | Welcome to the course | 287818 |
| 2 | 2 | 🎬 视频 | Overview of the AI models | Overview of the AI models | 287722 |
| 3 | 3 | 🎬 视频 | Accessing the API | Accessing the API | 287726 |
| 4 | 3 | 📄 图文 | Getting an API key | Getting an API key | 296766 |
| 5 | 3 | 🎬 视频 | Making a request | Making a request | 287725 |
| 6 | 3 | 🎬 视频 | Multi-Turn conversations | Multi-Turn conversations | 287733 |
| 7 | 3 | 🎬 视频 | Chat exercise | Chat exercise | 287727 |
| 8 | 3 | 🎬 视频 | System prompts | System prompts | 287734 |
| 9 | 3 | 🎬 视频 | System prompts exercise | System prompts exercise | 287724 |
| 10 | 3 | 🎬 视频 | Temperature | Temperature | 287728 |
| 11 | 3 | 📝 测验 | Course satisfaction survey | Course satisfaction survey | 297284 |
| 12 | 3 | 🎬 视频 | Response streaming | Response streaming | 287734b |
| 13 | 3 | 🎬 视频 | Structured data | Structured data | 287732 |
| 14 | 3 | 🎬 视频 | Structured data exercise | Structured data exercise | 287729 |
| 15 | 3 | 📝 测验 | Quiz on accessing the AI with the API | Quiz on accessing the AI with the API | 289117 |
| 16 | 4 | 🎬 视频 | Prompt evaluation | Prompt evaluation | 287731 |
| 17 | 4 | 🎬 视频 | A typical eval workflow | A typical eval workflow | 287736 |
| 18 | 4 | 🎬 视频 | Generating test datasets | Generating test datasets | 287737 |
| 19 | 4 | 🎬 视频 | Running the eval | Running the eval | 287738 |
| 20 | 4 | 🎬 视频 | Model based grading | Model based grading | 287739 |
| 21 | 4 | 🎬 视频 | Code based grading | Code based grading | 287740 |
| 22 | 4 | 🎬 视频 | Exercise on prompt evals | Exercise on prompt evals | 287741 |
| 23 | 4 | 📝 测验 | Quiz on prompt evaluation | Quiz on prompt evaluation | 289118 |
| 24 | 5 | 🎬 视频 | Prompt engineering | Prompt engineering | 287742 |
| 25 | 5 | 🎬 视频 | Being clear and direct | Being clear and direct | 287743 |
| 26 | 5 | 🎬 视频 | Being specific | Being specific | 287744 |
| 27 | 5 | 🎬 视频 | Structure with XML tags | Structure with XML tags | 287745 |
| 28 | 5 | 🎬 视频 | Providing examples | Providing examples | 287746 |
| 29 | 5 | 🎬 视频 | Exercise on prompting | Exercise on prompting | 287747 |
| 30 | 5 | 📝 测验 | Quiz on prompt engineering techniques | Quiz on prompt engineering techniques | 289119 |
| 31 | 6 | 🎬 视频 | Introducing tool use | Introducing tool use | 287748 |
| 32 | 6 | 🎬 视频 | Project overview | Project overview | 287749 |
| 33 | 6 | 🎬 视频 | Tool functions | Tool functions | 287750 |
| 34 | 6 | 🎬 视频 | Tool schemas | Tool schemas | 287751 |
| 35 | 6 | 🎬 视频 | Handling message blocks | Handling message blocks | 287752 |
| 36 | 6 | 🎬 视频 | Sending tool results | Sending tool results | 287753 |
| 37 | 6 | 🎬 视频 | Multi-turn conversations with tools | Multi-turn conversations with tools | 287754 |
| 38 | 6 | 🎬 视频 | Implementing multiple turns | Implementing multiple turns | 287755 |
| 39 | 6 | 🎬 视频 | Using multiple tools | Using multiple tools | 287756 |
| 40 | 6 | 🎬 视频 | Fine grained tool calling | Fine grained tool calling | 287757 |
| 41 | 6 | 🎬 视频 | The text edit tool | The text edit tool | 287758 |
| 42 | 6 | 🎬 视频 | The web search tool | The web search tool | 287759 |
| 43 | 6 | 📝 测验 | Quiz on tool use with the AI | Quiz on tool use with the AI | 289120 |
| 44 | 7 | 🎬 视频 | Introducing Retrieval Augmented Generation | Introducing Retrieval Augmented Generation | 287763 |
| 45 | 7 | 🎬 视频 | Text chunking strategies | Text chunking strategies | 287776 |
| 46 | 7 | 🎬 视频 | Text embeddings | Text embeddings | 287777 |
| 47 | 7 | 🎬 视频 | The full RAG flow | The full RAG flow | 287778 |
| 48 | 7 | 🎬 视频 | Implementing the RAG flow | Implementing the RAG flow | 287779 |
| 49 | 7 | 🎬 视频 | BM25 lexical search | BM25 lexical search | 287780 |
| 50 | 7 | 🎬 视频 | A Multi-Index RAG pipeline | A Multi-Index RAG pipeline | 287781 |
| 51 | 8 | 🎬 视频 | Extended thinking | Extended thinking | 287782 |
| 52 | 8 | 🎬 视频 | Image support | Image support | 287783 |
| 53 | 8 | 🎬 视频 | PDF support | PDF support | 287784 |
| 54 | 8 | 🎬 视频 | Citations | Citations | 287785 |
| 55 | 8 | 🎬 视频 | Prompt caching | Prompt caching | 287786 |
| 56 | 8 | 🎬 视频 | Rules of prompt caching | Rules of prompt caching | 287787 |
| 57 | 8 | 🎬 视频 | Prompt caching in action | Prompt caching in action | 287788 |
| 58 | 8 | 🎬 视频 | Code execution and the Files API | Code execution and the Files API | 287789 |
| 59 | 8 | 📝 测验 | Quiz on features of the AI | Quiz on features of the AI | 289121 |
| 60 | 9 | 🎬 视频 | Introducing MCP | Introducing MCP | 287790 |
| 61 | 9 | 🎬 视频 | MCP clients | MCP clients | 287791 |
| 62 | 9 | 🎬 视频 | Project setup | Project setup | 287792 |
| 63 | 9 | 🎬 视频 | Defining tools with MCP | Defining tools with MCP | 287793 |
| 64 | 9 | 🎬 视频 | The server inspector | The server inspector | 287794 |
| 65 | 9 | 🎬 视频 | Implementing a client | Implementing a client | 287795 |
| 66 | 9 | 🎬 视频 | Defining resources | Defining resources | 287796 |
| 67 | 9 | 🎬 视频 | Accessing resources | Accessing resources | 287797 |
| 68 | 9 | 🎬 视频 | Defining prompts | Defining prompts | 287798 |
| 69 | 9 | 🎬 视频 | Prompts in the client | Prompts in the client | 287799 |
| 70 | 9 | 🎬 视频 | MCP review | MCP review | 287800 |
| 71 | 9 | 📝 测验 | Quiz on Model Context Protocol | Quiz on Model Context Protocol | 289122 |
| 72 | 10 | 🎬 视频 | Anthropic apps | Anthropic apps | 287801 |
| 73 | 10 | 🎬 视频 | AI Code Assistant setup | AI Code Assistant setup | 287802 |
| 74 | 10 | 🎬 视频 | AI Code Assistant in action | AI Code Assistant in action | 287803 |
| 75 | 10 | 🎬 视频 | Enhancements with MCP servers | Enhancements with MCP servers | 287804 |
| 76 | 11 | 🎬 视频 | Agents and workflows | Agents and workflows | 287805 |
| 77 | 11 | 🎬 视频 | Parallelization workflows | Parallelization workflows | 287806 |
| 78 | 11 | 🎬 视频 | Chaining workflows | Chaining workflows | 287807 |
| 79 | 11 | 🎬 视频 | Routing workflows | Routing workflows | 287808 |
| 80 | 11 | 🎬 视频 | Agents and tools | Agents and tools | 287809 |
| 81 | 11 | 🎬 视频 | Environment inspection | Environment inspection | 287810 |
| 82 | 11 | 🎬 视频 | Workflows vs agents | Workflows vs agents | 287811 |
| 83 | 11 | 📝 测验 | Quiz on Agents and Workflows | Quiz on Agents and Workflows | 289123 |
| 84 | 12 | 📝 测验 | Final Assessment | Final Assessment | 287812 |
| 85 | 13 | 📄 图文 | Course Wrap Up | Course Wrap Up | 287813 |
🧭 章节要点与 CCAR-F 权重落点
| 章节 | 课时数 | 备考价值 | 备注 |
|---|---|---|---|
| 1 Introduction | 1 | — | 导览 |
| 2 the AI provider overview | 1 | D1 打底 | 模型谱系 |
| 3 Accessing the AI with the API | 13 | D3 + D5 | system prompt / temperature / 流式 / 结构化输出 |
| 4 Prompt evaluation | 8 | D3 | 评估工作流、model-based 与 code-based grading —— CCAR-F 里少见的系统讲评估 |
| 5 Prompt engineering techniques | 7 | D3 | XML 标签、示例、明确指令 —— 官方提示工程四件套 |
| 6 Tool use with the AI | 13 | D1 + D4 | schemas / results / 多轮 / fine grained / text edit / web search |
| 7 RAG and Agentic Search | 7 | D1 | 分块、嵌入、BM25、多索引 —— 检索架构 |
| 8 Features of the AI | 9 | D5 + D3 | 扩展思考、图像、PDF、引用、提示缓存、代码执行与 Files API |
| 9 Model Context Protocol | 12 | D4 | tools / resources / prompts 三原语完整实现 |
| 10 the AI provider apps | 4 | D2（弱） | AI Code Assistant 与 Computer Use |
| 11 Agents and workflows | 8 | D1 | 并行化 / 链式 / 路由 / workflows vs agents |
| 12 Final assessment | 1 | 自测 | 全课程结业测验 |
| 13 Wrapping up! | 1 | — | 结课 |
🎯 建议学习路线（按 CCAR-F 权重）
- §11 Agents and workflows ＋ §7 RAG（D1 27%）—— 最大权重，先建骨架
- §5 Prompt engineering ＋ §4 Prompt evaluation ＋ §3 的 system prompt/structured data（D3 20%）
- §6 Tool use ＋ §9 MCP（D4 18%）
- §8 的 Prompt caching（D5 15%）
- §10 只过一遍，D2 留给队列第 2 门《AI Code Assistant in Action》
⚠️ 与
课程003的重叠见 README「与已归档课程 003 的关系」表：§3/§6/§8/§9/§11 可直接复用 003 已译内容，本课真正的新增在 §4 评估、§5 提示工程、§7 RAG、§12 结业测验。
✅ 本课进度
- [x] 官方 About 全文抓取与精译
- [x] 13 章节 / 85 课时全景表（含 lesson ID 与直链）
- [x] CCAR-F 五大领域映射与学习路线
- [x] 建立生词表
30-单词表/04-考试核心生词与术语表.md（精讲 36 条 ＋ 速查 1071 条，按章节分组） - [x] 逐课精译：第 3–5、7、8、10、11 章已全部完成；第 6、9 章除测验外已完成（2026-10-27）
- [ ] 待补：第 1、2、12、13 章讲稿（共 4 课）；第 6、9 章与第 12 章 Final Assessment 测验（题目需登录态）
- [x] 逐课生词即时回灌表 04（2026-10-27 起按章节分组回灌，截至第 10 章）
📌 语料来源说明
About 与 Curriculum 取自官方 Skilljar 伙伴站公开课程页字段，英文原句未改措辞；中文为精译。课时正文（视频讲稿/字幕）需 Partner SSO 登录，见 40-认证事务/08-Partner SSO SOP。个人学习用途，CC BY-NC-SA 4.0。