这里不是第三方文章摘抄，而是 FDE FAN 自己的知识讲义。参考来源放在每页末尾，方便继续阅读。 These are FDE FAN teaching notes, not copied article excerpts. References are listed at the end of each page for further reading.
课程讲义 / AI-FDE Operating ModelTeaching Note / AI-FDE Operating Model
AI 飞轮里的 FDE：从部署者到生产系统编排者FDE in the AI Flywheel: From Deployer to Production Orchestrator
我们从 AI 生产飞轮理解 FDE：客户现场、业务上下文、模型能力、工具链和交付反馈不断循环。FDE 不是单纯写前端，也不是临时帮客户调模型，而是把需求、上下文、Agent、部署、评估和复盘连成可重复运转的系统。We understand FDE through the AI production flywheel: customer context, model capability, toolchain, deployment, and feedback reinforce one another. An FDE is not merely a front-end builder or prompt fixer, but the person who connects requirements, context, agents, deployment, evaluation, and retrospectives into a repeatable operating loop.
HFS Research · Accessed 2026-06-28课程讲义 / Context EngineeringTeaching Note / Context Engineering
当上下文被平台化：FDE 的角色如何升级When Context Becomes a Platform: How the FDE Role Changes
我们要理解企业为什么重新思考 FDE：过去大量依赖个人现场知识和手工集成的工作，正在被上下文平台、AI 原生工具和可复用组件重构。对课程来说，重点不是减少 FDE，而是训练 FDE 把一次性交付变成组织能力。We study why companies are rethinking FDEs: work once dependent on individual field knowledge and manual integration is being reorganized through context platforms, AI-native tools, and reusable components. The point is not to reduce FDEs, but to turn one-off delivery into organizational capability.
Unframe · Accessed 2026-06-28工具讲义 / Stack DesignTool Lesson / Stack Design
2026 FDE 工具栈：从 IDE 到遥测的交付链路The 2026 FDE Tool Stack: Delivery from IDE to Telemetry
我们把 FDE 工具栈拆成多条能力线：AI 编码、内部工具/低代码、工作流自动化、数据集成和可观测协作。工具不是炫技清单，而是把从需求到上线的每一段风险可视化。We break the FDE stack into capability lanes: AI coding, internal tools and low-code, workflow automation, data integration, and observability/collaboration. Tools are not a trophy list; they make risk visible across the path from requirement to launch.
Perspective AI · Accessed 2026-06-28角色讲义 / FDE RoleRole Lesson / FDE Role
什么是 Forward Deployed Engineer：角色边界与交付责任What Is a Forward Deployed Engineer: Role Boundaries and Delivery Accountability
我们先用岗位地图理解 FDE：FDE 位于客户、产品和工程之间，既要理解业务，又要把解决方案带到现场。我们把这个角色拆成训练标准：沟通、建模、实现、部署、文档和边界。We map the FDE role between customers, product, and engineering: understanding the business while taking solutions into the field. We translate this role into training standards: communication, modeling, implementation, deployment, documentation, and boundaries.
Invisible Technologies · Accessed 2026-06-28实践讲义 / AI UXPractice Lesson / AI UX
从 SIer 工作坊到 FDE：价值发现、NSM 与产物驱动提示From SIer Workshops to FDE: Value Discovery, NSM, and Artifact-Driven Prompting
我们把 FDE 式工作放进 SIer/UX 的日常训练：从客户访谈、工作坊、价值定义、NSM 设定，到用产物截图、流程图和已有资料提示 AI。前端 Demo 不是终点，而是让客户、模型和交付团队围绕同一个可见产物对齐。We place FDE-style work inside SIer and UX routines: customer interviews, workshops, value definition, NSM selection, and prompting AI with screenshots, flow maps, and existing artifacts. A front-end demo is not the endpoint; it is the shared artifact that aligns customer, model, and delivery team.
note.com / kamechi_ai_ux · Accessed 2026-06-28统一课程化框架Unified Course Framework
| 研究主题Research Theme | 对应训练Training Translation | 学生交付物Student Deliverable |
|---|---|---|
| AI flywheel | 需求、上下文、Agent、部署、复盘闭环Requirement, context, agent, deployment, retrospective loop | README / eval set / launch log |
| Context platform | 把一次交付变成可复用资产Turn one-off work into reusable assets | Prompt version / source map / permission notes |
| Tool stack | 按交付阶段选工具Select tools by delivery stage | Demo URL / logs / rollback note |