
The beginner's guide to coding with Cursor | Lee Robinson (Head of AI education)
Claire Vo (host), Lee Robinson (guest)
In this episode of How I AI, featuring Claire Vo and Lee Robinson, The beginner's guide to coding with Cursor | Lee Robinson (Head of AI education) explores cursor coding guide: agents, linting, rules, and better writing workflows Cursor is presented as an AI code editor that blends a traditional IDE with agentic workflows, letting users work from basic autocomplete up to an autonomous agent that can read files, run commands, and apply verified changes.
Cursor coding guide: agents, linting, rules, and better writing workflows
Cursor is presented as an AI code editor that blends a traditional IDE with agentic workflows, letting users work from basic autocomplete up to an autonomous agent that can read files, run commands, and apply verified changes.
Lee explains how adding classic software engineering “guardrails” (TypeScript/typed languages, linters, formatters, and tests) dramatically improves agent reliability because the AI can read failures and correct itself.
They demo an agent fixing lint errors end-to-end (run lint → identify issues → patch code → rerun lint to verify), then discuss running parallel tasks in the background while you keep coding.
The conversation also covers power features—@-mentioning Git context, custom rules and reusable commands like “code review,” model selection (Auto vs manual), chat/context management—and ends with a writing workflow that uses a “banned words/phrases” prompt to reduce AI-sounding copy.
Key Takeaways
Treat Cursor as a spectrum: autocomplete → agent.
Lee frames Cursor’s UI as progressive autonomy: basic inline suggestions, then higher-autonomy agent actions that can create files and make multi-step changes. ...
Get the full analysis with uListen AI
Add engineering guardrails to make agents consistently useful.
Typed languages, linters, formatters, and tests provide machine-readable feedback loops. ...
Get the full analysis with uListen AI
Use agents for closed-loop fixes: run → detect → patch → verify.
The lint demo shows the ideal agent pattern: it runs the lint command, applies fixes, and reruns lint to confirm success. ...
Get the full analysis with uListen AI
Always inspect diffs to learn and stay in control.
Claire emphasizes opening the red/green diff to understand what changed—especially for beginners. ...
Get the full analysis with uListen AI
Run parallel agent tasks to keep momentum.
Lee describes keeping focus in the main editor while delegating side tasks (e. ...
Get the full analysis with uListen AI
Codify recurring mistakes into custom rules and commands.
When a model repeats an undesired behavior, Lee recommends turning the correction into a reusable rule. ...
Get the full analysis with uListen AI
Use Auto model selection until you understand tradeoffs.
Auto reduces cognitive load by choosing a model based on speed/quality/availability. ...
Get the full analysis with uListen AI
Micro-slice chats to avoid context bloat and quality decay.
They note that long chats degrade as context fills; Cursor may summarize at the limit, but results can worsen near 80–90%. ...
Get the full analysis with uListen AI
Apply “linting” to writing to remove generic AI tone.
Lee’s writing mega-prompt bans hype words and common LLM patterns (e. ...
Get the full analysis with uListen AI
Notable Quotes
“Rather than giving you step-by-step instructions, you're just putting the thing into the GPS, and it just figures it out along the way.”
— Lee Robinson
“There are tools that you can take from traditional software engineering... and apply them to make your code more resilient to errors and help the AI models fix errors for you.”
— Lee Robinson
“You can't just vibe code your way forever.”
— Lee Robinson
“It's a linter and a formatter.”
— Claire Vo
“When I re-prompt... and I'm not very explicit about what actually wasn't good... the model can't read my mind.”
— Lee Robinson
Questions Answered in This Episode
In the lint-fixing demo, what specifically enabled the agent to choose `bun run lint`—was that inferred from package scripts, docs, or repo conventions?
Cursor is presented as an AI code editor that blends a traditional IDE with agentic workflows, letting users work from basic autocomplete up to an autonomous agent that can read files, run commands, and apply verified changes.
Get the full analysis with uListen AI
What are your minimum “guardrails” for a brand-new prototype (types, lint, format, tests)—and in what order should a beginner add them?
Lee explains how adding classic software engineering “guardrails” (TypeScript/typed languages, linters, formatters, and tests) dramatically improves agent reliability because the AI can read failures and correct itself.
Get the full analysis with uListen AI
Can you share the exact structure of your custom “code review” command prompt and how you iterated it over time based on real bugs?
They demo an agent fixing lint errors end-to-end (run lint → identify issues → patch code → rerun lint to verify), then discuss running parallel tasks in the background while you keep coding.
Get the full analysis with uListen AI
When should someone switch from Auto model selection to manual model choice in Cursor, and what tasks clearly benefit from a reasoning model?
The conversation also covers power features—@-mentioning Git context, custom rules and reusable commands like “code review,” model selection (Auto vs manual), chat/context management—and ends with a writing workflow that uses a “banned words/phrases” prompt to reduce AI-sounding copy.
Get the full analysis with uListen AI
What’s your recommended workflow for splitting work across multiple chats/agents without losing shared context (e.g., architecture decisions)?
Get the full analysis with uListen AI
Transcript Preview
What are some common tasks that you think Cursor can help people do that help you build better code, help you write better software?
In this instance, in this repo, I've already set up a linter, and there's something in my application that's not working correctly. So I can say, "Fix the lint errors." The nice thing about AI agents is they can go read things for you. So when I say, "Fix the lint errors," it knows that it can go run a terminal command, and it knows that, hey, there's this thing called bun run lint. So it went, and it ran the command bun run lint. It said there were two issues that it found, so it applies some code changes on this line, and then it reruns the same command, and it verifies itself that things were fixed. And I think this is a really interesting thing, which is that I didn't have to go tell the agent to do anything. It's like rather than giving you step-by-step instructions, you're just putting the thing into the GPS, and it just figures it out along the way. [upbeat music]
Welcome back to How I AI. I'm Claire Vo, product leader and AI obsessive, here on a mission to help you build better with these new tools. Today, I have a really fun episode with Lee Robinson, who's teaching the future of coding at Cursor. What I love about this episode is we're gonna walk through Cursor's AI agent and IDE, and we're gonna do it for the folks out there that are a little less technical. We're gonna show you how you can be less intimidated by code, learn some foundational concepts of software engineering, and use Cursor to ship more products. I really love this episode, and I hope you'll enjoy it, too. Let's get to it.
This podcast is supported by Google. Hey, everyone, Shresta here from Google DeepMind. The Gemini 2.5 family of models is now generally available. 2.5 Pro, our most advanced model, is great for reasoning over complex tasks. 2.5 Flash finds the sweet spot between performance and price, and 2.5 Flash Lite is ideal for low-latency, high-volume tasks. Start building in Google AI Studio at ai.dev.
Welcome to How I AI, Lee. I am so excited because you get to work with and on one of my daily driver AI products, Cursor, and so I'm excited personally to [chuckles] learn a couple of your tips and tricks and also to share them with the How I AI audience. So welcome.
Thank you for having me. I'm excited to be here.
As someone who I think also, uh, especially now, uses Cursor on a daily basis, one of the things I wanna zoom out and explain for people is, what the heck is Cursor? We, of course, have lots of sophisticated engineers as listeners, but we also have people that are new to building AI products and want to learn all these tools at their disposal, want to be able to build code and write software. And so where does Cursor sit in the ecosystem of, of tools for AI builders?
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