# Google 与 Kaggle 的 35.3 万人 vibe coding 课程回顾

- 来源：Google Blog：AI（RSS）
- 作者：Anant Nawalgaria
- 发布时间：2026-08-03 23:00
- AIHOT 分数：54
- AIHOT 链接：https://aihot.virxact.com/items/cmsesilco15d0ro2epbyhfuog
- 原文链接：https://blog.google/innovation-and-ai/technology/developers-tools/ai-agents-intensive-recap-2026

## AI 摘要

Google 与 Kaggle 的“AI Agents: Intensive Vibe Coding”课程吸引超 35.3 万人注册，聚焦通过自然语言编程的 vibe coding 范式。课程期间 Kaggle Discord 超 39.2 万参与者协作，收到 1.2 万名活跃学员提交的 6000 多份毕业项目。全部内容仍可通过 Kaggle Learn 自学获取。

## 正文

Google and Kaggle’s “AI Agents: Intensive Vibe Coding” course had expert-led sessions, technical whitepapers and notebooks, and capstone projects.

Anant Nawalgaria

Group AI Product Manager, Founder of GenAI intensive

Brenda Flynn

Partnerships Lead, Kaggle

AI is moving faster than traditional methods of learning and teaching can keep up with it. With new concepts, skills, ideas, and tools arriving almost daily, we needed a new way to teach and learn these new technologies.

Since Google and Kaggle’s first “5 Day Intensive” course in 2024 more than 2 million learners and developers have participated in the multimodal, no cost courses. Our latest collaboration, the 5-Day AI Agents: Intensive Vibe Coding Course with Google, had over 353,000 registered participants. Together, we explored the frontier of vibe coding, where people program through natural language. Our courses navigated the entire lifecycle of designing, securing, and deploying production-grade AI agents in the cloud. The response proved that developers are ready to graduate their prototypes "from vibe to live”.

Hundreds of thousands of learners gathered across codelabs and technical whitepapers to collaborate in real time. The heart of this collaboration was on Kaggle's Discord, where over 392,000 active participants swapped code, debugged together, and formed study groups.

The capstone projects brought the week’s learning into action. We received over 6,000 project submissions from more than 12,000 active capstone participants.

The projects demonstrated the immense creativity, technical skill, and imagination of the global AI developer community. From advanced digital humanities tools like the historical manuscript transcription pipeline Palimpsest, to sophisticated space-weather research systems like Project ARIES, the caliber of submissions made it incredibly difficult to narrow down the top entries. You can explore all the winning teams and their capstone projects on our website.

We look forward to continuing our partnership and exploring future learning opportunities. If you’re interested in learning more, you can continue honing your skills through ongoing Kaggle competitions and stay connected with fellow learners on the Kaggle Discord. For those who missed the live event in June but still want to learn, all course content remains available as a self-paced Kaggle Learn guide.