FDE-the-Guidance-Book-of-Forward-Deployed-Engineer README

The project's README from github.com/xdash/FDE-the-Guidance-Book-of-Forward-Deployed-Engineer. As one markdown file: README.md.

前线部署工程师:人工智能时代的客户价值交付秘籍

范冰 著 · 免费公开全文,欢迎在线阅读与分享

官网:fde4.ai(手机端适配、大陆访问更快;另有书外持续更新的案例库、生态地图、名录)


关于本书

2025 年夏天,我的朋友圈被同一个数字刷屏:95%。

麻省理工学院 NANDA 实验室 2025 年发布的《生成式人工智能的鸿沟》(The GenAI Divide) 报告说,过去三年,全球企业在生成式人工智能上烧了三四百亿美元,其中 95% 的项目没能产生任何能写进财务报表的价值。几乎同一时间,另一条新闻在往相反的方向狂奔:硅谷的招聘网站上,一个叫「前线部署工程师」(Forward Deployed Engineer,简称 FDE)的岗位,发布量一年涨了七倍多。OpenAI 在招,Anthropic 在招,YC 孵化器里一百多家创业公司都在招。

一边是企业人工智能项目 95% 的阵亡率,一边是一个岗位一年暴涨七倍多的抢手度。把这两条新闻摆在一起看,答案不难猜:模型已经不稀缺了,能把模型塞进客户真实业务里的人,才稀缺。

我由此对 FDE 这个岗位产生了浓厚的兴趣。多年前写《增长黑客》时,我做的事情本质上和这次一样:把一个硅谷正在发生、但国内还没有名字的东西,系统地研究一遍,再诚实地讲清楚。这次我也沿用了同样的笨办法——翻遍能找到的一手材料:Palantir 早期高管在播客里的复盘、前员工的回忆录、风投的行业分析、麻省理工那份报告的原文、几十份各家公司的招聘启事、薪酬报告、论坛上从业者的吐槽,以及国内第一批实践者的经验谈。

这本书就是那个研究过程的完整沉淀。它聊清楚三件事:

书里所有数据和案例,都在附录 C 里标明了出处。整理的过程本身就是学习,我尽量让每一条引用都经得起核查;如有疏漏,欢迎通过 Issue 指正。

为什么免费公开

这本书首先是我写给自己的学习笔记。研究一个陌生领域,最好的验收方式就是把它讲成一本别人也能读懂的书——讲不清楚的地方,就是自己没弄懂的地方。

整理完之后,把它锁在硬盘里,多少有些可惜。FDE 在国内还是一个刚有名字的岗位,很多讨论还停留在「它是不是换了个名字的售前」。如果这份沉淀能帮正在考虑转型的工程师看清这个岗位的全貌,帮投身企业人工智能的创业者少走一段弯路,帮任何一个想弄明白「人工智能到底怎么才能在企业里落地」的人节省几十个小时的检索时间,那它公开出来,显然比躺在我的硬盘里更有价值。

知识的价值在于流动,而不在于囤积。

关于作者

范冰,网名 XDash,互联网从业者,《增长黑客》作者,长期关注技术趋势与商业方法论。本书官网:fde4.ai。本书为个人研究整理,不代表任何机构立场。

联系方式:

阅读方式

目录

章节 内容
自序 为什么写这本书
第 1 章 FDE 的崛起
第 2 章 解决正确的问题
第 3 章 赢得客户
第 4 章 激活部署
第 5 章 守住续约
第 6 章 扩大收入
第 7 章 规模化复制
第 8 章 完整案例集
后记 FDE 的职业道德
附录 A FDE 应当关注的常用指标
附录 B FDE 人物与团队名单
附录 C 全书案例索引与资料出处

版权声明

本书著作权归作者范冰所有。本仓库内容由作者本人授权公开,供读者免费阅读与非商业性分享,转载请务必注明出处与作者;任何商业用途(包括但不限于出版、培训、付费内容改编)须事先获得作者书面许可。


如果这本书对你有帮助,欢迎 Star 本仓库,或把它分享给需要的人。

FDE-the-Guidance-Book-of-Forward-Deployed-Engineer (xdash/FDE-the-Guidance-Book-of-Forward-Deployed-Engineer)

Searching Database…

What is this?A free, full-text online book about the Forward Deployed Engineer (FDE) role in the AI era, written in Chinese by Fan Bing (XDash), author of 'Growth Hacker'. It is readable on the companion site and chapter by chapter on GitHub.
Is this one book, or a bigger collection of resources?One book of eight chapters plus three appendices. Its companion site fde4.ai also hosts extras beyond the book: a continuously updated case library, an ecosystem map, and a directory.
How do I read it?Three ways: read on the official site fde4.ai (mobile-friendly, kept in sync with the repo), open chapters directly on GitHub via the table of contents, or download the merged PDF (v1.0.24). For updates, the chapter Markdown is authoritative.
Who is this book for?Engineers considering a move into the FDE role, founders building enterprise AI who want to avoid detours, and anyone trying to understand how AI actually delivers value inside companies.
What exactly is a Forward Deployed Engineer?An engineer who gets AI models into real customer businesses — a role that grew out of Palantir's intelligence projects. Demand is exploding: postings grew over 7x in a year, with OpenAI, Anthropic and 100+ YC startups hiring, while MIT found 95% of enterprise GenAI projects deliver no measurable value.
What does the book actually cover?Three things: what the FDE role is and why it boomed (Ch. 1); how the job is done across a full delivery journey — finding the right problem, winning customers, activating deployments, keeping renewals, expanding revenue, scaling (Ch. 2-7); and who does it, via 165 real cases from Palantir, OpenAI, Anthropic, Harvey and Sierra (Ch. 8).
Why did the author make the book free?It began as his own study notes — he argues the best test of understanding a field is explaining it in a book others can read. Since the FDE role is still new in China, he chose to publish openly rather than leave it on his hard drive.
Are the cases and figures verifiable?Yes — every data point and case cites its source in Appendix C, and the author invites readers to flag any mistakes through GitHub Issues. Appendix B also lists notable FDE people and teams.
Creator
xdash
Size
26.4 MB
Stars
5.1K
Updated
CommitAug 14
Age
2mo
Commits · 90 Days