来源： Latent Space播客 | Cat Wu（Claude Code产品经理）、Boris Cherny（Claude Code首席工程师） | 2025年5月7日 分类： Anthropic 原文发表： May 07, 2025 纪要生成： 2026-02-25
本次播客邀请到Anthropic旗下Claude Code项目的产品经理Cat Wu和首席工程师Boris Cherny，深度拆解这款终端编程智能体的诞生背景、产品设计哲学和核心功能亮点。两位嘉宾结合Anthropic内部的使用实践，分享了AI编程工具的技术选型思路、成本收益逻辑，以及大模型时代开发者工作流的变革方向。同时披露了Claude Code的迭代路线、商业化规划，以及Anthropic面向开发者的产品布局思路。
本节重点 - Claude Code最初是Boris的个人实验项目，无预先规划，因内部使用率爆发式增长才正式立项 - 项目初期仅3人核心团队，Cat因高频使用并输出大量反馈被邀请加入担任PM - 核心定位是运行在终端的AI智能体，可直接访问本地文件、执行bash命令，支持完全自主工作流
详细精要
💬 精华片段（中文）
"Claude Code根本不算一个产品，它更像是一个Unix工具。"
"Claude Code is not a product as much as it’s a Unix utility."
本节重点 - 核心产品原则是「先做最简单的事」，优先用最少资源验证产品市场匹配度，再逐步扩张 - PM采用轻量级管理模式，大部分功能需求来自内部用户的真实使用反馈，极少顶层规划 - 产品迭代以3个月为周期对齐大模型能力迭代节奏，优先适配未来模型可支撑的场景
详细精要
💬 精华片段（中文）
"我们不做Cursor或者Windsurf这类产品，虽然它们非常优秀，很多人每天都在用，我自己也用。我们想做的是处在曲线更早期的产品，随着模型能力提升，未来一年甚至更长时间会成长为大规模产品的东西。"
"We would build, you know, a cursor or a wind serve or something like this. Like, these are awesome products that so many people use every day. I use them. That's not the product that we want to build. We want to build something that's kind of much earlier on that curve and something that will maybe be a big product, you know, a year from now or, you know, however much time from now."
本节重点 - 功能实现分为三层：模型内置、Claude Code脚手架层、外部工具组合层，优先选择最简化的实现路径 - 仅当功能无法由模型或外部工具实现时，才会内置到Claude Code的脚手架中 - 上下文压缩功能最终采用让Claude自主总结历史消息的方案，无需复杂工程开发即可满足需求
详细精要
💬 精华片段（中文）
"当模型足够优秀的时候，最简单的方案通常都能生效，你完全不需要过度工程。"
"And it's funny when the model is so good, the simple thing usually works. You don't have to over-engineer it."
本节重点 - 放弃复杂的记忆架构方案，采用本地markdown文件Claude.md作为用户自定义记忆载体 - 支持根目录、子目录、home目录多位置存放Claude.md，系统会自动读取对应文件到上下文 - 核心逻辑是用最简单的可用方案满足记忆需求，降低用户理解和使用成本
详细精要
💬 精华片段（中文）
"我们有很多关于记忆架构的疯狂想法，业内也有大量相关研究和外部产品，我们也从这些内容中获得了灵感，但最终我们还是推出了最简单的方案：就是一个存放内容的文件，会被自动读取到上下文里。"
"We had all these crazy ideas about like memory architectures and, you know, there's so much literature about this. There's so many different external products about this and we wanted to be inspired by all this stuff. But in the end, the thing we did is ship the simplest thing, which is, you know, it's a file that has some stuff. And it's auto-read into context."
本节重点 - 定位为面向高阶用户的「功率工具」，核心优势是提供模型的原生访问能力，无上层UI封装的额外开销 - 适合批量自动化任务场景，例如同时启动上千个实例批量修复Lint错误、生成PR等 - 和Cursor、Devin等产品形成互补，而非直接竞争，Anthropic内部也在同时使用各类AI编程工具
详细精要
💬 精华片段（中文）
"如果你想要一个能直接访问模型、可以用Claude自动化大规模工作负载的功率工具，比如你有上千个Lint错误，想要启动上千个Claude实例逐个修复然后生成PR，那么Claude Code是非常合适的工具。它是面向高阶用户、高阶工作负载的工具。"
"So if you want to use a power tool that lets you access the model directly and use Claude for automating, you know, big workloads, you know, for example, if you have like a thousand Lint violations and you want to start a thousand instances of Claude and have it fix each one and then make a PR, then ClaudeCode is a pretty good tool. Got it. It's a tool for power workloads for power users."
本节重点 - 核心设计理念对齐Unix工具哲学，支持和其他命令行工具灵活组合，嵌入任意工作流 - 内部用户已有单日消耗上千美元的自动化用例，适合大规模并行任务场景 - 非交互模式支持完全自动化运行，无需人工介入即可完成批量任务
详细精要
💬 精华片段（中文）
"我们把它看作一个Unix工具，就像你组合grep、cat或者其他工具一样，你可以把Code组合到你的工作流里。"
"We think of it as like a Unix utility. Mm-hmm. Right? So it's like the same way that you would compose, you know, grep or cat or, oh, cat. Or something like this. Nice. The same way you can compose code into workflows."
本节重点 - 目前采用按token付费的模式，单活跃用户日均花费约6美元，高于Cursor每月20美元的订阅制成本，但ROI更突出 - 成本和latency强相关，团队优先保障工具响应速度和任务完成的完整度，其次才是成本优化 - 核心价值逻辑是ROI优先：工程师人力成本很高，哪怕仅提升50%生产力，对应的价值也远高于工具成本
详细精要
💬 精华片段（中文）
"我认为这是一个ROI问题，而非成本问题。想想工程师的平均薪资，如果能让工程师的生产力提升50%到70%，那对应的价值是非常高的，这才是正确的思考方式。"
"I would add that I think the way I think about it is it's an ROI question. It's not a cost question. And so if you think about, you know, an average engineer salary and like what, you know, we were talking about this before the podcast. Like, engineers are very expensive. And if you can make an engineer 50, 70% more productive, that's worth a lot. And I think that's the way to think about it."
本节重点 - 核心上线功能包括网页抓取、自动补全、自动上下文压缩、自动接受、Vim模式、自定义斜杠命令、标签记忆等 - 网页抓取功能做了严格的安全限制，仅抓取用户明确提供的链接或已抓取页面内的链接，保障企业使用安全 - 自动接受功能响应了用户对高信任场景下自主运行的需求，支持模型自主编辑文件、运行测试，无需人工逐次确认
详细精要
💬 精华片段（中文）
"我们注意到很多用户表示「我已经非常信任Claude Code了，我想让它自主编辑我的文件、运行测试，完成后再回来找我」，所以我们推出了自动接受功能。"
"We also shipped auto accept because we noticed that a lot of users were like, hey, like Claude Code can figure it out. I've like developed a lot of trust for Claude Code. I wanted to just like autonomously edit my files, run tests, and then come back to me later. So those are some of the big ones."
本节重点 - Claude Code约80%-90%的代码由自己生成，仅复杂的数据模型重构等场景需要人工编写 - 工作流为Claude先生成代码，人工审核后合并，大幅降低了开发工作量 - 这类AI辅助开发模式已经成为行业普遍现象，多家A轮阶段公司的AI生成代码占比也达到80%-85%
详细精要
💬 精华片段（中文）
"通常我们的工作流是Claude先写代码，如果效果不好，再由人工介入。还有一些场景我更倾向于自己手写，比如复杂的数据模型重构，因为我有很明确的设计想法，直接实现比给Claude解释要更快。"
"So usually where we start is quad writes the code. And then if it's not good, then maybe a human will dive in. There's also some stuff where like I actually prefer to do it by hand. So it's like, you know, intricate data model refactoring or something. I won't leave it to quad because I have really strong opinions and it's easier to just do it and experiment than it is to explain it to quad."
本节重点 - 自定义斜杠命令本质是保存的prompt，适合简单的本地个性化需求，无需使用MCP - Claude Code同时支持MCP客户端和MCP服务器，可灵活组合不同工具的能力 - 典型用例为通过自定义斜杠命令实现语义Lint，结合GitHub MCP自动提交PR修复问题
详细精要
💬 精华片段（中文）
"我们认为用户不应该被绑定到某一种特定技术上，应该用最适合自己需求的方案。"
"We think generally you shouldn't have to be tied to a particular technology. You should use whatever works for you."
本节重点 - 技术栈采用Ink（React终端渲染框架）+ Bun（JavaScript运行时），大幅提升终端开发效率 - 终端开发面临跨终端兼容性问题，类似早年浏览器兼容问题，Ink很好地抽象了这层差异 - 权限系统允许用户自定义允许/禁止的操作，默认允许读文件，编辑、运行命令等操作可按需配置白名单
详细精要
💬 精华片段（中文）
"用Ink开发终端UI有点像早年开发浏览器应用，需要考虑IE6、Opera、Firefox等不同浏览器的兼容性，每个终端的ANSI实现都有一点差异，Ink很好地帮我们抽象了这层差异。"
"So building in this way, it feels to me a little bit like building for the browser back in the day where you had to think about like Internet Explorer 6 versus Opera versus like Firefox and whatever. Like you have to think about these cross-terminal differences a lot. Yeah. So yeah, big fans of Ink because it helps abstract over that."
本节重点 - Claude Code作为底层原语，支持用户自主构建代码审查、安全扫描、语义Lint等上层工具 - 语义Lint可覆盖传统规则型Lint无法覆盖的场景，比如代码与注释一致性、业务规则校验等 - 企业引入AI生成代码后，代码审查的核心逻辑不变，仍由提交代码的工程师对最终代码质量负责
详细精要
💬 精华片段（中文）
"以前我在别人PR下评论「能不能补一下测试」会觉得很不好意思，因为大家都知道写测试很麻烦，很多人会为了赶进度跳过。但现在我都会直接提，因为Claude可以直接写好测试，不需要人工付出额外工作量。"
"And, you know, before, I felt like a jerk if on someone's PR, I'm like, hey, can you write a test? Because, you know, they kind of know they want to… For code coverage? Is that still relevant? For code coverage, yeah. Okay. And, you know, they kind of know they should probably write a test and that's probably the right thing to do. And somewhere in their head, they make that trade-off where they just want to ship faster. And so you always kind of feel like a jerk for asking. But now I always ask because Claude can just write the test."
本节重点
- 非交互模式通过claude -p "prompt"调用，适合自动化批量任务场景
- 最佳实践是先从只读任务开始测试，逐步放开权限，小范围验证后再扩大规模
- 典型企业用例包括批量修复过时/不稳定测试、生成变更日志、批量更新文档等
详细精要
-p参数传入prompt即可启动非交互模式，无需人工介入--allow-tools参数指定允许使用的工具，比如仅允许读文件、仅允许git相关命令等，限制操作权限💬 精华片段（中文）
"很多使用Claude Code的企业都会用非交互模式，比如他们会说「我的代码库中有几十万个测试，有些过时了，有些不稳定」，然后让Claude Code逐个查看这些测试，决定如何更新、是否需要废弃，提升代码覆盖率。"
"And also a lot of our, the companies using quad code actually use this non-interactive mode. So they'll, for example, say, hey, I have, like, hundreds of thousands of tests in my repo. Some of them are out of date. Some of them are flaky. And they'll send quad code. So they'll send quad code to look at each of these tests and decide, okay, how can I update any of them? Like, should I deprecate some of them? How do I, like, increase our code coverage? So that's been a really cool way that people are non-interactively using quad code."
本节重点 - 核心度量指标包括：周期时间（Cycle Time）、原本不会开发的功能数量 - 传统的代码行数、PR数量等指标存在缺陷，但仍是目前可用的最不坏的度量方式 - 代码覆盖率、类型覆盖率、圈复杂度等代码质量指标仍有参考价值，可根据团队特性选择使用
详细精要
💬 精华片段（中文）
"我们从客户反馈中发现了一个很常见的模式：客服或客户成功团队反馈一个小bug，10分钟后对应团队的工程师就说「Claude Code已经做好修复了」。很多工程师都表示，如果没有Claude Code，他们根本不会花时间修复这类小问题，因为会打断当前的工作节奏，最终只会积压在待办列表里。"
"We have a lot of channels where we get customer feedback. And one of the patterns that we've seen with Claude Code is that sometimes customer support or customer success will, like, post, hey, like, this app has, like, this bug. And then sometimes 10 minutes later, one of the engineers on that team will be, like, Claude Code made a fix for it. And a lot of those situations when you, like, ping them and you're, like, hey, that was really cool, they were, like, yeah, without Claude Code, I probably wouldn't have done that because it would have been too long. It would have been too much of a divergence from what I was otherwise going to do. It would have just ended up in this long backlog."
本节重点 - 新功能仍保持很高的准入门槛，要求直观易用、无额外学习成本、符合产品整体愿景 - AI降低了原型开发成本，现在可以快速实现多个版本的原型，通过实际试用验证方案，而非仅靠设计文档讨论 - 内部工具、零到一的原型开发是Claude Code的核心优势场景，大幅降低了这类需求的开发门槛
详细精要
💬 精华片段（中文）
"以前我会先写一份很长的设计文档，花很多时间思考问题，再开始开发。现在我会直接让Claude Code写3个不同版本的原型，试用后看哪个我更喜欢，这比写文档能更快、更好地帮我做出决策。"
"Where, like Cat's saying, like, before I would write a big design doc. And I would think about a problem for a long time before I would build it sometimes for some set of problems. And now I'll just ask Claude Code to prototype, like, three versions of it. And I'll try the feature and see which one I like better. And then that informs me much better and much faster than a doc would have."
本节重点 - 记忆功能目前仍在探索阶段，已观察到用户的创新用法，比如让Claude写操作日志，记录团队工作习惯、目标、工作风格等 - 早期曾尝试RAG实现代码库检索，最终替换为智能体搜索方案，效果更好且无索引同步、安全等问题 - 团队认为随着模型能力提升，模型最终会内置自己的知识存储能力，外部记忆工具的价值会逐渐降低
详细精要
💬 精华片段（中文）
"我加入Anthropic之前肯定会说知识图谱是记忆的最优解，但现在我觉得一切都是模型，最终获胜的一定是模型能力的提升。随着模型越来越好，它会覆盖所有其他组件的能力，模型会自己编码知识图谱、自己编码KV存储，你只需要给它合适的工具就行。"
"Are you a believer in knowledge graphs for this stuff? You know, I'm a big, if you talked to me before I joined Anthropic and this team, I would have said, yeah, definitely. But now, actually, I feel everything is the model. Like, that's the thing that wins in the end. And it just, as the model gets better, it subsumes everything else. So, you know, at some point, the model will encode its own knowledge graph. It'll encode its own, like, KV store if you just give it the right tools."
本节重点 - 规划功能没有单独的模式，用户直接通过自然语言要求Claude思考、制定计划即可，无需额外的功能入口 - 沙箱、环境分支等能力可通过外部工具组合实现，Claude Code作为底层原语无需内置 - 目前模型的主要缺陷包括：过于严格遵循字面需求忽略隐含意图、长会话中容易丢失原始上下文
详细精要
💬 精华片段（中文）
"最新的Sonnet 3.7是非常执着的模型，非常有动力完成用户的目标，但有时候会把用户的目标理解得过于字面，不会去满足请求中隐含的部分，因为它太专注于「我必须完成X」。我们正在想办法给它注入更多常识，让它知道努力和用户不想要的行为之间的边界。"
"I think one of the things about the latest Sonnet 3.7 is it's a very persistent model. It's like very, very motivated to accomplish the user's goals. But it sometimes takes the user's goal very literally. And so it doesn't always fulfill what like the implied parts of the request are, because it's just so narrowed in on like, I must get X done. And so we're trying to figure out, okay, how do we give it a bit more common sense so that it knows the line between trying very hard and like, no, the user definitely doesn't want that."
本节重点 - Claude Code已经成为Anthropic的长期正式项目，团队正在持续扩张，不会停止迭代 - 目前采用按token付费模式，未来会探索订阅制方案，满足用户对成本可预测性的需求 - 正在评估开源/源码开放的可能性，核心障碍是开源社区的维护成本较高 - 生产力提升的实测数据仍在收集中，内部实测平均提升2x，部分工程师可达10x
详细精要
💬 精华片段（中文）
"对我个人来说，Claude Code大概提升了2倍的生产力。Anthropic内部有些工程师的生产力提升甚至达到10倍，也有一些用户还没找到合适的使用方式，仅用来生成提交信息，提升大概10%，所以整体的提升范围差异很大，我们还在做更严谨的研究。"
"But anecdotally for me, it's probably 2x my productivity. My God. So I'm just like, I'm an engineer that codes all day, every day. Yeah. For me, it's probably 2x. Yeah. I think there's some engineers at Anthropic where it's probably 10x their productivity. And then there's some people that haven't really figured out how to use it yet. And, you know, they just use it to generate like commit messages or something. That's maybe like 10%. So I think there's probably a big range and I think we need to study more."
本节重点 - 核心基础是Claude模型本身的编码能力很强，是Claude Code等开发者工具的核心支撑 - 开发者工具领域的ROI非常清晰，提升开发者生产力可以直接释放巨大的经济价值 - 内部团队充分吃自己的狗粮，所有模型团队都在使用Claude Code，能快速发现模型缺陷并迭代优化
详细精要
💬 精华片段（中文）
"我觉得很多情况是模型本身就很会写代码，我们是站在 incredible 团队的肩膀上，这是Claude Code存在的唯一原因。"
"It's like, I feel like the model just wants to write code. Yeah, I think a lot of this trickles down from like the model itself being very good at code generation. Like we're very much building off the backs of an incredible team. I think that's the only reason why Claude Code is possible."
| 术语 | 解释 |
|---|---|
| Claude Code | Anthropic推出的运行在终端的编程智能体，可直接访问本地文件、执行bash命令，提供Claude模型的原生访问能力 |
| Sonnet | Anthropic旗下的大模型系列，是Claude Code默认使用的模型，在编码能力和速度之间实现了较好的平衡 |
| Haiku | Anthropic旗下的轻量级大模型，速度快成本低，适合预提交检查、简单规则校验等轻量任务 |
| MCP（Model Control Protocol） | Anthropic推出的模型控制协议，支持不同AI工具之间的能力互通和组合 |
| Unix Utility | 遵循Unix设计哲学的命令行工具，核心特点是轻量、单一职责、可灵活组合 |
| Bitter Lesson（苦涩的教训） | AI领域的经典观点，认为通用计算和算力的提升最终会超过针对特定场景的人工优化，本集中Anthropic的产品设计遵循该原则，优先做通用、底层的能力 |
| Cyclomatic Complexity（圈复杂度） | 衡量代码复杂度的指标，本集中提到是目前相对有效的代码质量度量标准之一 |
| RAG（Retrieval Augmented Generation） | 检索增强生成技术，本集中提到Claude Code早期曾尝试使用RAG实现代码库检索，最终替换为智能体搜索方案 |
| ASL（Autonomous Safety Level） | Anthropic内部的自主安全等级体系，用于衡量AI系统自主运行时的安全程度 |
| Ink | 基于React的终端UI渲染框架，Claude Code使用该框架开发终端交互界面 |
| Bun | 高性能JavaScript运行时，Claude Code使用Bun进行代码编译和测试运行 |
| Tmux | 终端多路复用工具，用户可使用该工具和Claude Code组合实现多窗口并行工作流 |
| Husky | Git预提交钩子工具，用户可通过该工具将Claude Code集成到代码提交前的检查流程 |
| Aider | 较早推出的CLI-based AI编程工具，是Claude Code的同类竞品 |
| Cursor | 获得90亿美元估值的AI IDE产品，是AI编程工具赛道的主流玩家之一 |
| Windsurf | 被OpenAI以30亿美元收购的AI IDE产品 |
| Chain of Thought（思维链） | 一种大模型提示技术，要求模型逐步输出推理过程，提升输出结果的准确性 |