Introducing ShipAI

How do you know an AI project is real? You know it’s real when you watch it run. That’s the idea behind ShipAI, which is live on TDS today. It’s a video showcase for AI work: practitioners record a screen-share walkthrough of something they’ve built (an app, an agent, a pipeline, an experiment), and we review and publish it. 你怎么知道一个 AI 项目是真实的？当你亲眼看到它运行时，你就知道它是真的。这就是 ShipAI 背后的理念，它今天已在 TDS 上线。这是一个展示 AI 作品的视频平台：从业者录制一段屏幕共享演示，展示他们构建的内容（应用程序、智能体、流水线或实验），我们负责审核并发布。

What’s on a project page

项目页面包含什么

Every entry is a video walkthrough (4-15 minutes long) in which the creator explains what inspired their project, how they built it, how it works, and what they’ve learned along the way. Around the video you’ll find: 每个条目都是一段视频演示（时长 4-15 分钟），创作者在视频中解释其项目的灵感来源、构建方式、工作原理以及在此过程中学到的经验。在视频周围，你还会发现：

• AI-generated key takeaways, so you can decide in ten seconds whether to press play. AI 生成的关键要点，让你能在十秒钟内决定是否点击播放。
• A searchable transcript, lightly cleaned, so you can jump straight to the part you need. 可搜索的文字记录（经过轻微整理），让你能直接跳转到需要的部分。
• Stack notes, including the models, infrastructure, and tools in the build, linked out. 技术栈说明，包括构建中使用的模型、基础设施和工具，并附带外部链接。
• Links to where the project lives, whether on GitHub, Hugging Face, or a live demo. We showcase the work; we don’t host it. 项目地址链接，无论是 GitHub、Hugging Face 还是在线演示。我们负责展示作品，但不托管代码。

Every builder also gets a profile page collecting everything they’ve shipped. 每位开发者还会拥有一个个人主页，汇总他们发布过的所有作品。

What you’ll find on day one

首日上线内容

ShipAI opens with more than 30 published walkthroughs recorded over the summer by our founding builders, most of them TDS authors whose work you may already know. We already published a retrieval experiment on RAG chunk sizes, an AI damage-report generator for warehouse operations, and a CV turned into a playable video game. ShipAI 上线首日即发布了 30 多段演示视频，这些视频由我们的创始开发者在夏季录制，其中大多数是你们可能已经熟悉的 TDS 作者。我们已经发布了关于 RAG 分块大小的检索实验、用于仓库运营的 AI 损坏报告生成器，以及一个被改造成可玩电子游戏的简历。

Some are weekend experiments, while others are already running in production: we showcase projects across a wide range of development stages. 有些是周末的实验性项目，而另一些已经在生产环境中运行：我们展示了处于不同开发阶段的各类项目。

An editor reviews every submission

编辑审核每一份提交

The same editorial team that reviews TDS articles also runs ShipAI. We don’t prioritize production value and flashy visuals. Instead, we choose videos that help other builders understand what you made, how, and why. 负责审核 TDS 文章的编辑团队同样负责运营 ShipAI。我们并不优先考虑制作水准和华丽的视觉效果，而是选择那些能帮助其他开发者理解你做了什么、如何做以及为什么这么做的视频。

Show us what you built

展示你的作品

Have you shipped something real recently? We want to see it! Record a screen walkthrough (using Loom, OBS, QuickTime, or any other tool that captures your screen and your voice) and send us the link through the submission form. 你最近发布了什么实际成果吗？我们想看看！录制一段屏幕演示（使用 Loom、OBS、QuickTime 或任何其他能捕捉屏幕和声音的工具），并通过提交表单将链接发送给我们。

ShipAI is now live at towardsdatascience.com/shipai. Go watch what people are building. And if one of the walkthroughs leaves you thinking “I should share my project, too,” that’s just what we’re after. ShipAI 现已在 towardsdatascience.com/shipai 上线。去看看大家都在构建什么吧。如果其中某段演示让你产生“我也应该分享我的项目”的想法，那正是我们所期待的。

Questions about a submission, or anything else? Email us at shipai@towardsdatascience.com. 对提交内容或其他方面有疑问？请发送电子邮件至 shipai@towardsdatascience.com。