【文章标题】:I spent $20,000 on Devin in a month. Here’s what I learned | Ryan Carson (solo founder) 【文章标题】:我在一个月里花了 20,000 美元在 Devin 上。以下是我学到的东西 | Ryan Carson(独立创始人)

【文章正文】: 【文章正文】:

Ryan Carson Ryan Carson

is a five-time founder and the current solo founder of Untangle, a B2B SaaS platform for family law firms. Before Untangle, he co-founded Treehouse, an online coding education platform, and has spent the better part of two decades building and leading tech companies. He’s active on X, where he shares his solo founder journey in real time, including what he actually spends on AI tools each month. 他是五次创业者,目前是 Untangle 的独立创始人,Untangle 是一个面向家事法律事务所的 B2B SaaS 平台。在 Untangle 之前,他联合创立了在线编程教育平台 Treehouse,并花了近二十年的大部分时间构建和领导科技公司。他活跃于 X(推特),实时分享他的独立创始人旅程,包括他每月实际在 AI 工具上的花费。

Listen or watch on

YouTube

,

Spotify

, or

Apple Podcasts 在 YouTube、Spotify 或 Apple Podcasts 上收听或观看

What you’ll learn: 你将学到:

  • Why Ryan manages 15 concurrent Devin agents with a folder system and a piece of paper, not a dashboard

  • 为什么 Ryan 用文件夹系统和一张纸(而不是仪表盘)管理 15 个并发的 Devin 智能体

  • The Watchdog playbook: what he built to replace a customer success team across every law firm account

  • Watchdog 操作手册:他构建了什么来取代每个律师事务所账户的客户成功团队

  • How his LAN PR skill closes the loop on 40 daily PRs without a QA team reviewing a single one

  • 他的 LAN PR 技能如何在没有任何 QA 团队审查的情况下闭环处理每天 40 个 PR

  • Why he moved off local agents almost entirely, and the one situation where he still reaches for Codex

  • 为什么他几乎完全放弃了本地智能体,以及他仍然使用 Codex 的唯一场景

  • The design workflow we’re both using: Claude Design into a Markdown spec, then Codex to build the real thing

  • 我们都在使用的设计工作流:先用 Claude Design 生成 Markdown 规格说明,然后用 Codex 构建真实产品

  • What he found when he got away from his computer and met a real customer, and why it changed his entire product direction

  • 当他离开电脑去见真实客户时发现了什么,以及为什么这改变了他的整个产品方向

  • How he’s hiring his first engineer without a single phone screen or interview

  • 他如何在没有一次电话筛选或面试的情况下招聘第一位工程师

  • Why we both think more AI output is actually the wrong goal, and what to optimize for instead

  • 为什么我们都认为更多的 AI 产出实际上是错误的目标,以及应该优化什么

Brought to you by: 本期由以下赞助商提供:

WorkOS

—Make your app enterprise-ready, with SSO, SCIM, RBAC, and more WorkOS — 让你的应用具备企业级能力,支持 SSO、SCIM、RBAC 等

Jira AI SDLC

—Get your tokens’ worth with Jira Jira AI SDLC — 用 Jira 让你的 token 物有所值

In this episode, we cover: 在本期节目中,我们涵盖:

(00:00) Introduction to Ryan Carson (00:00) Ryan Carson 介绍

(02:55) Ryan’s update: Untangle, the divorce PMF pivot, and B2B growth (02:55) Ryan 的最新进展:Untangle、离婚产品市场契合点转型以及 B2B 增长

(07:35) Ryan’s current Devin stack: folders, P0 threads, and the paper list (07:35) Ryan 当前的 Devin 技术栈:文件夹、P0 线程和纸质清单

(16:29) Watchdog playbook: account monitoring across every firm (16:29) Watchdog 操作手册:跨每个律所账户的监控

(18:09) Managing agent decision fatigue: Ryan’s method vs. Claire’s (18:09) 管理智能体决策疲劳:Ryan 的方法 vs. Claire 的方法

(20:32) Producing more output does not make a better product (20:32) 产出更多并不等于产品更好

(22:47) Using cloud agents for ops beyond just code (22:47) 将云端智能体用于代码以外的运营工作

(25:14) When to use Codex vs. Devin vs. Claude Code (25:14) 何时使用 Codex、Devin 与 Claude Code

(27:15) Merge Mommy recap (27:15) Merge Mommy 回顾

(28:15) LAN PR skill: review loops, video walkthrough, and auto-merge (28:15) LAN PR 技能:审查循环、视频演示和自动合并

(30:20) Slack vs. Devin threads for async team communication (30:20) 用于异步团队沟通的 Slack 与 Devin 线程对比

(35:58) Claude Design plus Codex for building a technical design system (35:58) 使用 Claude Design 加 Codex 构建技术设计系统

(39:00) EA tools: Polly the OpenClaw vs. Claude Code on a Mac Mini (39:00) EA 工具:OpenClaw 上的 Polly 与 Mac Mini 上的 Claude Code

(42:09) Quick recap and final thoughts (42:09) 快速回顾与最终思考

Tools referenced: 参考工具:

• Devin (Cognition): https://www.cognition.ai/ • Devin (Cognition):https://www.cognition.ai/

• Codex (OpenAI): https://openai.com/codex • Codex (OpenAI):https://openai.com/codex

• Claude Code/Claude Design (Anthropic): https://www.anthropic.com/claude • Claude Code/Claude Design (Anthropic):https://www.anthropic.com/claude

• OpenClaw (Claude-based desktop client): https://openclaw.ai • OpenClaw(基于 Claude 的桌面客户端):https://openclaw.ai

• Cursor: https://www.cursor.com/ • Cursor:https://www.cursor.com/

• BugBot (Devin’s built-in PR review): https://cursor.com/bugbot • BugBot(Devin 内置的 PR 审查):https://cursor.com/bugbot

• Sentry (error monitoring referenced in Watchdog): https://sentry.io/ • Sentry(Watchdog 中引用的错误监控):https://sentry.io/

• Ugmonk (analog to-do system): https://ugmonk.com/ • Ugmonk(模拟待办事项系统):https://ugmonk.com/

Other references: 其他参考:

• Devin playbooks/skills documentation: https://docs.cognition.ai/ • Devin 操作手册/技能文档:https://docs.cognition.ai/

• Jack Dorsey/Buzz (async-first communication referenced): https://buzz.new/ • Jack Dorsey/Buzz(引用的异步优先沟通):https://buzz.new/

• Merge Mommy (Claire’s Eve agent for PR risk scoring, deployed on Vercel): https://www.lennysnewsletter.com/p/build-an-ai-code-review-bot-in-30 • Merge Mommy(Claire 的 Eve 智能体,用于 PR 风险评分,部署在 Vercel 上):https://www.lennysnewsletter.com/p/build-an-ai-code-review-bot-in-30

Where to find Ryan Carson: 在哪里找到 Ryan Carson:

X: https://x.com/ryancarson X:https://x.com/ryancarson

Untangle: https://untangle.us Untangle:https://untangle.us

Where to find Claire Vo: 在哪里找到 Claire Vo:

ChatPRD: https://www.chatprd.ai/ ChatPRD:https://www.chatprd.ai/

Website: https://clairevo.com/ 网站:https://clairevo.com/

LinkedIn: https://www.linkedin.com/in/clairevo/ LinkedIn:https://www.linkedin.com/in/clairevo/

X: https://x.com/clairevo X:https://x.com/clairevo

Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email jordan@penname.co. 制作与营销由 https://penname.co/ 负责。如需咨询播客赞助事宜,请发送邮件至 jordan@penname.co。