Claire Vo · 3 min read · Gen AI Tools
I spent $20,000 on Devin in a month. Here’s what I learned | Ryan Carson (solo founder)
Read the original on lennysnewsletter.com →
Ryan Carson breaks down what he learned from spending $20k per month on Devin, including a lightweight operational stack of 15 agents managed with folders and a paper list and a Watchdog system to monitor multiple accounts. He walks through a Claude Design into Markdown spec workflow, when to use Codex vs Devin, and how real customer interactions redirected Untangle’s product direction. He also shares pragmatic hiring steps and why optimizing AI output, not simply increasing it, matters.
Key Takeaways
- Devin can be effectively managed with simple systems (folders, P0 threads, and a paper to-do list) even at scale (15 agents) without dashboards.
- The Watchdog playbook enables cross-account monitoring across many firms, serving as a proxy for a customer success function.
- A practical design workflow using Claude Design to Markdown specs, then Codex to build, can streamline AI-assisted product development.
- Direct customer engagement away from the computer can pivot product direction in meaningful ways.
- Hiring and governance approaches—such as hiring the first engineer without traditional phone screens—can align teams with AI-driven output goals that prioritize quality over quantity.
Excerpt
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. Listen or watch on YouTube , Spotify , or Apple Podcasts What you’ll learn: Why Ryan manages 15 concurrent Devin agents with a folder system and a piece of paper, not…