scripod.com

人类和 AI Agent 的最佳配合方式,还没被发明|对谈 Paperboy

Shownote

本周,十字路口的嘉宾是 Paperboy(https://www.paperboy.com)团队。John Yang 21 岁,CEO。Jett Chen 19 岁,CMU 大一在读,同时是 founding engineer。Paperboy 团队 12 人,10 位工程师,融资 470 万美金。 John 认为:人类和 AI Agent 配合工作的最佳方式,很可能还没被发明出来。虽然已经有了 Claude Code、Codex、Manus、OpenClaw,但它们本质上都是 session-base...

Highlights

This podcast episode features the Paperboy team, including 21-year-old CEO John Yang and 19-year-old founding engineer Jett Chen, who are rethinking how humans and AI agents collaborate. They argue that current session-based, prompt-driven tools like Claude Code and Manus are fundamentally flawed, and they are building a new kind of agent that learns continuously from user activity.
00:00
Paperboy's mission is to explore the best way to collaborate with AI
01:59
The optimal human-AI collaboration method hasn't been invented yet.
06:24
Current tools lack a new dimension because they are still limited by human teams.
08:01
Paperboy solves this by having agents learn from observing user activity
14:21
Proactive agent design remains an underexplored domain
22:00
Agent predicts user intent without explicit instructions
30:19
Slack has strong network effects making it hard to replace
33:46
Different time horizons require different product forms for automation.
44:38
College provides time to build technical skills and explore.
52:21
The biggest company will come from consumer AI, not enterprise.

Chapters

快问快答
00:00
起点:现在的 AI 产品,我用着不爽
01:59
Claude Code 轰炸之下
06:21
Agent 的两大问题
08:01
屏幕数据成了行业共识之后
14:21
Mini Vivian & Auto John
16:46
微信群聊,启发了界面设计
27:36
The Last Interface 与五种速度
33:36
两种工程师、一本书、一个教练
42:09
拒绝过 Cognition、Vercel、Sentry,然后呢
48:07

Transcript

Koji: 嗨,我是 Koji,那本周十字路口呢?我们请到的嘉宾是 Paperboy 的创始人 John Yang,以及他们的一位 Founding Engineer Jett Chen。你好二位,欢迎来到十字路口。 John Yang: Hello Koji。 Koji: 在这里也给大家说明一下,因为江用英文表达会比较自然,所以我们这一期内容会比较中音夹杂。我们同时有音频和视频,如果大家这个需要的话,可以去看视频,看视频呢,我们就会配字幕和字幕的翻译,帮助大家理解。我们还是从十字路口的传统,快问快答开始。...
小宇宙
Open in 小宇宙