scripod.com

“I haven’t written a single line of front-end code in 3 months”: How Notion’s design team uses Claude Code to prototype

How I AI

Feb 23
How I AI

How I AI

Feb 23

Shownote

Brian Lovin is a designer at Notion AI who has transformed how the design team builds prototypes, by creating a shared code environment powered by Claude Code. Instead of designers working in isolated repositories or limited to static Figma designs, Brian built a collaborative “prototype playground” where the entire team can create, share, and iterate on functional prototypes. In this episode, Brian demonstrates how AI-assisted coding has dramatically accelerated the design process and why code-based prototyping is essential for building AI-powered products. What you’ll learn: 1. How Brian built a shared Next.js app that serves as a collaborative prototyping environment for Notion’s design team 2. Why encountering “reality” early in the design process leads to better products 3. How to use Claude Code’s “plan mode” to get better results when prototyping 4. The power of custom Claude slash commands and skills to automate repetitive tasks 5. How to transform Figma designs into working code with a single prompt 6. Why AI-powered products can’t be effectively designed in static tools like Figma 7. Brian’s rule for working with AI: “When Claude asks you to do something, teach it to do that thing itself” — Brought to you by: WorkOS [https://workos.com/?utm_source=lennys_howiai&utm_medium=podcast&utm_campaign=q22025]—Make your app enterprise-ready today Orkes [https://www.orkes.io/]—The enterprise platform for reliable applications and agentic workflows — In this episode, we cover: (00:00) Introduction to Brian (02:36) Building for B2B SaaS (04:42) Notion’s prototype playground: what it is and how it works (08:01) The technical background of designers using the playground (10:52) Demo: building a podcast player prototype (16:00) Actionable tips for better Claude Code results (20:16) Analyzing the result (20:30) Creating slash commands to simplify the workflow (23:03) Turning Figma designs into production-ready code (25:06) MCP frustrations and tips (30:54) Demo: creating a custom “find icon” skill (35:03) Demo: Creating a deploy command to simplify GitHub workflows (41:09) Quick recap (41:59) How code-based prototyping is changing design at Notion (46:48) Brian’s tool preferences (48:42) Prompting techniques when AI is not listening — Tools referenced: • Claude Code: https://claude.ai/ [https://claude.ai/] • Cursor: https://cursor.sh/ [https://cursor.sh/] • Next.js: https://nextjs.org/ [https://nextjs.org/] • Figma: https://figma.com/ [https://figma.com/] • Monologue: https://www.monologue.to/ [https://www.monologue.to/] • GitHub: https://github.com/ [https://github.com/] • GitHub Desktop: https://desktop.github.com/ [https://desktop.github.com/] • Tailwind CSS: https://tailwindcss.com/ [https://tailwindcss.com/] • Bun: https://bun.sh/ [https://bun.sh/] — Other references: • Claude Skills explained: How to create reusable AI workflows: https://www.lennysnewsletter.com/p/claude-skills-explained [https://www.lennysnewsletter.com/p/claude-skills-explained] — Where to find Brian Lovin: Website: https://brianlovin.com/ [https://brianlovin.com/] LinkedIn: linkedin.com/in/brianlovin [http://linkedin.com/in/brianlovin] X: https://twitter.com/brian_lovin [https://twitter.com/brian_lovin] — Where to find Claire Vo: ChatPRD: https://www.chatprd.ai/ [https://www.chatprd.ai/] Website: https://clairevo.com/ [https://clairevo.com/] LinkedIn: https://www.linkedin.com/in/clairevo/ [https://www.linkedin.com/in/clairevo/] X: https://x.com/clairevo [https://x.com/clairevo] — Production and marketing by https://penname.co/ [https://penname.co/]. For inquiries about sponsoring the podcast, email jordan@penname.co.

Highlights

Brian Lovin, a designer at Notion AI, discusses how he reimagined design collaboration by building a shared, code-based prototyping environment powered by AI—specifically Claude Code.
00:00
AI coding tools like Claude Code enable faster and more efficient prototyping
02:37
AI coding tools let designers prototype faster and more realistically by encountering design problems earlier
04:42
Designers can namespace directories for prototypes in Notion's Next.js playground
08:01
Some non-technical people made their first AI-assisted prototypes in the playground
13:40
Using plan mode before coding leads to better outcomes
19:08
Teach Claude to run ESLint and access MCP tools to self-check instead of relying on browser verification
20:24
The power of MCP is amazing
20:32
Typing '/create prototype' can start the creation process
23:03
The /figma command automates Figma-to-Notion UI implementation with ~80% accuracy
25:06
Pasting a Figma file link builds a working design with minimal setup
30:54
'Find icon' skill searches for icons using synonyms and related words to avoid hallucinated names
40:46
Enabled auto-merge in a somewhat risky way (Yolo)
41:12
The key rule is to teach Claude to do tasks when it asks.
44:54
Designers must prototype with actual code to see model performance and design plausible products
46:48
Each AI model has a unique 'mouth feel' reflecting its subjective user experience and interaction style
48:43
Good prompts depend on the quality of context and one's state

Chapters

Introduction to Brian
00:00
Building for B2B SaaS
02:36
Notion’s prototype playground: what it is and how it works
04:42
The technical background of designers using the playground
08:01
Demo: building a podcast player prototype
10:52
Actionable tips for better Claude Code results
16:00
Analyzing the result
20:16
Creating slash commands to simplify the workflow
20:30
Turning Figma designs into production-ready code
23:03
MCP frustrations and tips
25:06
Demo: creating a custom “find icon” skill
30:54
Demo: Creating a deploy command to simplify GitHub workflows
35:03
Quick recap
41:09
How code-based prototyping is changing design at Notion
41:59
Brian’s tool preferences
46:48
Prompting techniques when AI is not listening
48:42

Transcript

Brian Lovin: The way I think about designing B2B SaaS is you want your designs to encounter reality as early as possible. I've always been into prototyping, and then all of a sudden these AI coding tools come along. And now I can prototype faster, I can pr...