Aakash Gupta

Give me 60 minutes, I'll make your AI Designing 81% Better

Xinran Ma breaks down the complete AI design workflow. Mind map of AI design, live demos in Google AI Studio and Lovable, plus the exact tools top designers use. Full Writeup: https://www.news.aakashg.com/p/xinran-ma-podcast Transcript: https://www.aakashg.com/designing-ai-products-the-right-way-google-stitch-custom-gpts-and-prototyping-workflows-with-xinran/ --- Timestamps: 0:00 - Intro 3:54 - What Is Designing with AI? 8:02 - The AI Design Mind Map 12:29 - Ads 14:00 - Custom GPT to PRD Workflow 26:10 - Building Custom GPT Live Demo 30:02 - Ads 32:25 - Generating PRD for Prototyping 36:07 - Comparing Lovable vs V0 vs Bolt 41:23 - Stitch to AI Studio Workflow 43:31 - Google Stitch Live Demo 49:53 - YOLO Mode for Divergent Solutions 56:06 - Advanced Google AI Studio Tips 57:32 - How Cursor Stacks Up 59:37 - Final Takeaways 1:01:00 - Outro ---- 🏆 Thanks to our sponsors: 1. NayaOne: The fastest way to test AI and fintech solutions - https://nayaone.com/ 2. Pendo: The #1 software experience management platform - http://www.pendo.io/aakash 3. Maven: Get 15% off Xinran’s course with my link - https://bit.ly/3Y2FUZn 4. Bolt: Ship AI-powered products 10x faster - https://bolt.new/solutions/product-manager?utm_source=Promoted&utm_medium=email&utm_campaign=aakash-product-growth 5. Gamma: Turn customer feedback into product decisions with AI - https://gamma.app/?utm_campaign=prompt&utm_content=Aakash+Gupta&utm_source=LinkedIn --- Key Takeaways: 1. AI design covers five areas not just prompts - Prompting, ideation, design/prototyping, workflows, and staying conscious. Most people think better prompts equal better design. That's just 20% of the skill. 2. Use Google AI Studio for quick design variations - Upload 2-3 visual references. Describe what you want. Generate three different design directions in 5 minutes. What used to take 3-4 hours now takes 15 minutes. 3. Lovable builds functional prototypes in seconds - Describe the experience you want to build. Lovable generates a working prototype in 60 seconds. Not mockups—actual clickable experiences you can test with users. 4. Match tools to specific use cases - Custom GPT for effective prompts. Lovable for high-quality prototypes. Magic Patterns for design variations. Google AI Studio for free exploration. Cursor for full-stack experiences. Claude Code as all-purpose best. 5. Good design passes four layers not just visual - Visual representation, problem-solving, design principles, and implementation feasibility. Most people stop at layer one. Great design works at all four layers. 6. Context matters more than prompt length - Don't say "design a button." Say "design a primary CTA button for B2B SaaS onboarding where users connect calendar. Professional brand." Specificity drives quality. 7. Visual references anchor AI output - Upload 2-4 screenshots showing the aesthetic you want. These show AI what "modern and minimal" means to you. The quality difference is massive versus text-only prompts. 8. Iteration speed determines final quality - The magic isn't in the first output. It's in the 10th iteration after you've refined and tweaked. Review, identify issues, tell AI how to fix, repeat. 9. Always validate with real users - AI tools make generating designs easy. Only users tell you if those designs actually help. Show prototypes to 3-5 users. Watch them try to use it. 10. Workflows changed from linear to parallel - Before AI: sequential steps taking weeks. After AI: describe, generate, iterate freely. This is how top 1% designers work now. --- 👨‍💻 Where to find Xinran Ma: LinkedIn: https://www.linkedin.com/in/davidmaxinran/ Newsletter: https://www.designwithai.co/ 👨‍💻 Where to find Aakash: Twitter: https://www.x.com/aakashg0 LinkedIn: https://www.linkedin.com/in/aagupta/ Newsletter: https://www.news.aakashg.com #aidesign #productmanagement --- 🧠 About Product Growth: The world's largest podcast focused solely on product + growth, with over 200K+ listeners. 🔔 Subscribe and turn on notifications to get more videos like this.

AGAakash GuptahostXMXinran Maguest

Feb 20, 2026·1h 1m·Watch on YouTube ↗

📖 CHAPTERS

1. 1

   Why PMs get AI product design wrong: workflows over prompts

   Aakash frames the core problem: many PMs treat “designing with AI” as just writing prompts, which produces shallow, generic outcomes. Xinran positions the right mental model as understanding end-to-end workflows, constraints, and how tools behave in practice.

2. 2

   What “Designing with AI” includes: the 4-part landscape

   Xinran introduces a mind map that broadens “AI design” beyond tactics. She breaks the space into prompting, ideation, design/prototyping workflows, and a less-discussed area: conscious/intentional design with AI and risk awareness.

3. 3

   Prompting fundamentals for design: clarify ask, context, and references

   Instead of piling on frameworks, Xinran simplifies prompting into the minimum that reliably improves design outputs. She emphasizes clarifying the request, providing only necessary context, and adding references that shape structure and quality.

4. 4

   How specific should prompts be? Balancing control vs exploration

   Aakash probes the tradeoff between specificity (control) and openness (divergence). Xinran describes prompting as an ‘art’—like delegating to a teammate—where early structure helps but you sometimes leave room for creativity to avoid over-constraining ideas.

5. 5

   Ideation and prototyping with AI: divergent + convergent thinking

   Xinran explains how ideation and prototyping blend together with modern tools. She highlights giving AI guardrails for brainstorming, then forcing convergence via ranking, evidence, examples, and evaluation criteria—especially because AI struggles with priorities.

6. 6

   Designing consciously: risks, empathy gaps, and human-in-the-loop safeguards

   Xinran adds the ethical and quality layer: AI can hallucinate, generalize, and miss human nuance. She argues intentionality and empathy matter more as AI makes generating artifacts cheap—so teams must validate inputs and retain human judgment.

7. 7

   Workflow 1 overview: Custom GPT → “PRD for prototyping” → any builder

   Xinran introduces a repeatable workflow for turning fuzzy ideas into a clean, tool-ready spec. The goal isn’t a full PRD—it’s a lightweight front-end-focused spec that can be pasted into prototyping tools for consistent results.

8. 8

   Building the Custom GPT logic: the questions that make the first prompt strong

   Aakash asks what to put into the GPT instructions so others can replicate it. Xinran explains the key intake questions: user, needs, goal, and—critically—what specific flow/platform to prototype first to keep AI focused and prevent scope blowups.

9. 9

   Live demo: generating a lightweight spec and sanity-checking in Claude Artifacts

   Xinran runs the Custom GPT to create an expense-tracking prototype spec, then uses Claude as a quick “mock run” to validate the prompt visually. She explains why Claude is useful as a fast preview layer even if it’s not the best design generator.

10. 10

    Comparing prototyping tools: Lovable vs V0 vs Bolt (and where Magic Patterns/Replit fit)

    Xinran compares popular tools on design quality, editing accessibility, and feature depth. She notes these are no longer simple “Claude wrappers”—tools add layers (agents/system prompts) that materially improve polish and interactions.

11. 11

    Workflow 2 overview: Stitch for design exploration → Google AI Studio for interaction

    Xinran introduces a newer combo workflow: Stitch for early design ideation and divergence, then Google AI Studio for interactive prototyping. The pairing is positioned as ‘best of both’: concept exploration plus runnable behavior.

12. 12

    Stitch live demo: redesigning an existing UX with YOLO-mode divergence

    Using a Redfin ‘Ask Redfin’ AI chat section as the reference, Xinran shows how Stitch can generate multiple redesign options. She demonstrates variation controls—especially YOLO mode—to push divergent layouts, styles, and content treatments.

13. 13

    Stitch → Google AI Studio handoff: turning a concept into an interactive prototype

    Xinran exports a chosen Stitch design to Google AI Studio to add interactivity. She notes a practical limitation: exporting to Figma may require specific modes (e.g., ‘fast’), and richer prototypes often require generating multi-screen flows, not just a single page.

14. 14

    Advanced Google AI Studio tips + Cursor comparison + final takeaways

    Xinran shares two AI Studio power moves: adding a hidden system instruction and using ‘Annotate App’ for comment-driven iteration. She then compares Cursor as a more flexible but harder-to-learn path, and Aakash summarizes the two main workflows as “AI superpowers.”