A 3-step AI coding workflow for solo founders | Ryan Carson (5x founder)
Ryan Carson (guest), Claire Vo (host)
In this episode of How I AI, featuring Ryan Carson and Claire Vo, A 3-step AI coding workflow for solo founders | Ryan Carson (5x founder) explores ryan Carson’s structured AI coding workflow: PRDs, tasks, context control Ryan Carson demonstrates a lightweight but structured approach to AI-assisted coding in Cursor: generate a PRD, convert it into a detailed task list, then execute tasks one subtask at a time with explicit “stop and confirm” checkpoints.
Ryan Carson’s structured AI coding workflow: PRDs, tasks, context control
Ryan Carson demonstrates a lightweight but structured approach to AI-assisted coding in Cursor: generate a PRD, convert it into a detailed task list, then execute tasks one subtask at a time with explicit “stop and confirm” checkpoints.
The central message is that most AI coding failures come from rushing context—slowing down to provide clear requirements and a plan actually speeds development and reduces rabbit holes and reverts.
He shows how Cursor “rules” (reusable prompt files) enforce consistent outputs (junior-dev-friendly PRDs, Markdown task lists with checkboxes, and task-by-task execution discipline).
Beyond the core workflow, he highlights MCP servers (notably Postgres) to reduce engineering toil and Repo Prompt to precisely package repo context for larger-model deep analysis outside Cursor’s context “black box.”
Key Takeaways
Don’t rush context—requirements clarity is the real accelerator.
Carson argues the most common mistake is impatience: skipping the time to explain what the AI needs to know. ...
Get the full analysis with uListen AI
Use a PRD prompt that’s “junior developer implementable.”
Framing the PRD for a junior developer forces the model to spell out assumptions, steps, and edge cases that a “genius PhD” model might otherwise skip, improving implementation fidelity.
Get the full analysis with uListen AI
Convert PRDs into checklisted tasks before writing code.
A structured task list (Markdown, checkboxes, subtasks/sub-subtasks) turns vague intent into executable steps and helps the AI and human stay aligned on scope and sequencing.
Get the full analysis with uListen AI
Force the AI to do one subtask at a time with explicit stops.
His task-management rule makes the agent complete a single subtask, mark it done, then ask to proceed—preventing runaway changes across the codebase and making review/revert easier.
Get the full analysis with uListen AI
Keep planning lightweight—Markdown can beat heavy tooling.
Even though he considered automating tasks via Asana/MCP, he prefers a hand-cranked Markdown task list for visibility, editability, and lower operational overhead.
Get the full analysis with uListen AI
MCPs reduce tab-switching toil; Postgres is the highest leverage.
He uses a Postgres MCP to ask natural-language questions about live data without writing SQL, and views MCPs as a path to putting browser, DB, and task execution into one interface.
Get the full analysis with uListen AI
When context must be exact, export curated repo context explicitly.
Repo Prompt helps select only relevant folders/files, tracks token counts, and emits a structured prompt (XML-tagged files) for models like o3—useful when Cursor’s context selection feels opaque.
Get the full analysis with uListen AI
Notable Quotes
“I think the biggest mistake that I do, that everyone does, is they try to rush through the context.”
— Ryan Carson
“If we all just slow down a tiny bit and do these two steps, it speeds everything up.”
— Ryan Carson
“This is a PRD that’s suitable for a junior developer to understand and implement this feature.”
— Ryan Carson
“Nobody really knows how to do this stuff. The only way you're really gonna figure it out is by getting in here and getting your hands dirty and see what works.”
— Ryan Carson
“Building this new startup, I literally feel like I'm able to do all of it… But I am able, for sure, to build this company by myself.”
— Ryan Carson
Questions Answered in This Episode
In your PRD rule, which sections are non-negotiable vs optional depending on feature size (e.g., non-goals, edge cases, success metrics)?
Ryan Carson demonstrates a lightweight but structured approach to AI-assisted coding in Cursor: generate a PRD, convert it into a detailed task list, then execute tasks one subtask at a time with explicit “stop and confirm” checkpoints.
Get the full analysis with uListen AI
Your task generator asks clarifying questions and then waits for a “Go”—what are the most common clarifications you find materially improve the resulting task plan?
The central message is that most AI coding failures come from rushing context—slowing down to provide clear requirements and a plan actually speeds development and reduces rabbit holes and reverts.
Get the full analysis with uListen AI
How do you decide the granularity of subtasks so the agent doesn’t overreach but you also don’t create a 200-item checklist?
He shows how Cursor “rules” (reusable prompt files) enforce consistent outputs (junior-dev-friendly PRDs, Markdown task lists with checkboxes, and task-by-task execution discipline).
Get the full analysis with uListen AI
What are the most frequent failure modes you still see even with the PRD → tasks → subtask execution workflow (lint errors, wrong file edits, missed edge cases, scope creep)?
Beyond the core workflow, he highlights MCP servers (notably Postgres) to reduce engineering toil and Repo Prompt to precisely package repo context for larger-model deep analysis outside Cursor’s context “black box.”
Get the full analysis with uListen AI
You mentioned keeping a “relevant files” list at the top—how do you maintain that list over time, and does it measurably reduce mis-edits by the model?
Get the full analysis with uListen AI
Transcript Preview
I think the biggest mistake that I do, that everyone does, is they try to rush through the context, where you just don't have the patience to tell the AI what it actually needs to know to solve your problem. And I think if we all just slow down a tiny bit and do these two steps, it speeds everything up. Nobody really knows how to do this stuff. The only way you're really gonna figure it out is by getting in here and getting your hands dirty and see what works.
That's a place where so many engineers and product managers get stuck in a loop. Like, who's gonna take this PRD and actually break it down in the right steps? So even just this is such a time-saver for people building products.
Building this new startup, I literally feel like I'm able to do all of it. Am I able to do it as well as a dedicated product manager? No. Am I able to think as deeply as a CTO? No. But I am able, for sure, to build this company.
This is the way, people. I'm telling you, pay attention. [upbeat music] Welcome to How I AI. I'm Claire, product leader and AI obsessive, here on a mission to help you build better with these new tools. 2025 is definitely the year of the vibe coder, but you can't always vibe your way to a scalable execution strategy. In this episode, Ryan Carson, a five-time founder with 20 years experience, shows us how he brings PRDs, task lists, and some advanced prompting techniques to Cursor to make sure he's not just vibing, he's building the right things. Let's get to it. Today's episode is brought to you by ChatPRD. I know that many of you are tuning in to How I AI to learn practical ways you can apply AI and make it easier to build. That's exactly why I built ChatPRD. ChatPRD is an AI copilot that helps you write great product docs, automate tedious coordination work, and get strategic coaching from an expert AI CPO. And it's loved by everyone, from the fastest-growing AI startups to large enterprises with hundreds of PMs. Whether you're trying to vibe code a prototype, teach a first-time PM the ropes, or scale efficiently in a large organization, ChatPRD helps you do better work fast. And we're integrated with the tools you love: v0.dev, Google Drive, Slack, Linear, Confluence, and more, so you don't have to change your workflow to accelerate with AI. Try ChatPRD free at chatprd.ai/howiai, and let's make product fun again. Hey, Ryan, it's nice to have you here!
Thanks. It's exciting to be here. I've listened to every episode so far, and I'm honored to be here myself. I can't wait.
So I'm gonna start with an easy question, which is: What are the last three things you built with AI?
I don't know if you call constantly using ChatGPT with your kids building something in AI, [chuckles] but I feel like, I feel like I'm the constant AI coach in our family, and I'm always delighted, actually, with what our kids are doing. And because of that, uh, my amazing 14-year-old kiddo, Devin, he said, "Dad, like, you know, I've been thinking about this game..." And I said, "Well, like, let's build it." And so we're building a, a primitive little side-scroller, and he's, like, the creative director, so, uh-
Install uListen to search the full transcript and get AI-powered insights
Get Full TranscriptGet more from every podcast
AI summaries, searchable transcripts, and fact-checking. Free forever.
Add to Chrome