Skip to content
Aakash GuptaAakash Gupta

If you can’t AI prototype after this, nothing will help you

Sachin Rekhi (Former Head of Product of LinkedIn Sales Navigator) breaks down the complete AI prototyping system. The 15-skill mastery ladder, live demos, and why Anthropic builds features this way. Full Writeup: https://www.news.aakashg.com/p/sachin-rekhi-podcast Transcript: https://www.aakashg.com/ai-prototyping-mastery-sachin-rekhi/ ---- Timestamps: 0:00 - Intro 0:40 - How Anthropic Builds Product Differently 3:36 - The Problem: AI Slop 8:41 - The AI Prototyping Mastery Ladder 11:38 - Design Consistency & Baselining 16:04 - Ad 17:03 - Diverging: The Secret Weapon 29:43 - Making Prototypes Functional 30:09 Ad 31:13 - Magic Patterns Demo 39:47 - Customer Validation Techniques 48:52 - When to Use Workflows vs Agents 57:00 - AI Prototyping Tools Face-Off 1:09:38 - Outro ---- 🏆 Thanks to our sponsor: Reforge: AI prototyping built for product teams - https://reforge.com/aakash ---- Key Takeaways: 1. Product shaping changes everything - Anthropic builds multiple prototypes for every problem, launches internally, sees what people use, then productionizes winners. This used to only be possible at Apple with massive labs. 2. AI slop is real - Type "create a CRM" and you get generic styling, vanilla features, basic scenarios. Looks magical but you'd never ship it. The challenge is going from slop to production-grade prototypes. 3. The 15-skill mastery ladder - Apprentice level: prompting, editing, design consistency. Journeyman: versioning, debugging, diverging. Master: functional prototyping, product shaping, analytics integration. 4. Design consistency starts with baselining - Take screenshot of your product. Recreate it. Iterate until perfect. Save as template. Now every prototype inherits your design system automatically. 5. Diverging is the secret weapon - Generate 4 design variants instead of 1. Magic Patterns has this built in. Or use multiple tools to get 8 options. Evaluate alternatives like designers do. 6. Functional prototypes unlock real validation - Integrate OpenAI API for actual responses. Add PostHog for session recordings and heatmaps. Build surveys. Track clicks. Test with real data, not mockups. 7. The tools face-off: which to actually use - Bolt for speed. V0 for beautiful UIs. Replit for full-stack. Magic Patterns for product teams with diverging. Reforge Build for context integration. Cursor for technical PMs. 8. The $5/month unlimited execution hack - Host n8n on Hostinger instead of paying per execution. Get unlimited runs. Build workflow that backs up to Google Drive for version history. 9. PMs can build what used to require engineering - Calendar integration. Email agents. Analytics dashboards. Multi-model comparison. Survey collection. All from prompts. No code required. 10. Traditional workflows beat agents for production - Workflows save tokens, run faster, and are more reliable. Use agents only when tasks need real decision-making. For known processes, use workflows. ---- 👨‍💻 Where to find Sachin Rekhi: LinkedIn: https://www.linkedin.com/in/sachinrekhi/ Newsletter: https://www.sachinrekhi.com/ Reforge AI Prototyping Course: https://reforge.com/Aakash 👨‍💻 Where to find Aakash: Twitter: https://www.x.com/aakashg0 LinkedIn: https://www.linkedin.com/in/aagupta/ Newsletter: https://www.news.aakashg.com #aiprototyping #aipm ---- 🧠 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.

Aakash GuptahostSachin Rekhiguest
Jan 25, 20261h 12mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

From AI slop to high-craft prototypes with validated solutions fast

  1. Anthropic’s approach flips traditional roadmapping by prototyping many problem-solution pairs first, dogfooding them, and only then productionizing the best-performing prototypes.
  2. “AI slop” happens when prototypes are generic, undifferentiated, and shallow in real workflows, but high-craft outcomes are achievable with the right techniques.
  3. Rekhi’s AI Prototyping Mastery Ladder outlines 15 skills from apprentice (prompting/editing/design consistency) to journeyman (debugging/versioning/diverging/validation) to master (functional prototypes and product shaping).
  4. Core methods include baselining your existing product via screenshot recreation, iterating with targeted edits (including batching), and forking templates to ensure consistent design across prototypes.
  5. Master-level validation uses deployed prototypes with embedded surveys plus analytics (e.g., PostHog events, heatmaps, session replays) to scale learning and simplify interfaces based on real behavior.

IDEAS WORTH REMEMBERING

5 ideas

Prototype many problem-solution pairs before committing roadmap capacity.

Anthropic-style “product shaping” prioritizes solutions that are already internally or customer-vetted, reducing the risk of building the wrong thing even when the problem is real.

Treat “one-shot apps” as a starting point, not a shippable output.

The first AI-generated UI is often generic in styling, undifferentiated in concept, and weak on real user workflows—use it to accelerate iteration, not to declare victory.

Baseline your product’s look-and-feel, then build everything on top of it.

Recreating a screenshot, refining it through edits, and duplicating/forking that baseline lets future prototypes inherit components and styling automatically, eliminating “wireframe AI” aesthetics.

Batch related edits to reduce round-trip time—without losing control.

Grouping similar changes (e.g., multiple color tweaks) speeds iteration, but batching unrelated changes makes it harder to isolate failures and manage versions when something breaks.

Use diverging intentionally—and use multiple tools to expand the idea space.

Ask for multiple variants (and even run the same explore prompt in different tools) because differing system prompts yield meaningfully different designs, producing more inspiration than a single tool run.

WORDS WORTH SAVING

5 quotes

They’re prioritizing not only what is actually… a problem worth solving, but a problem-solution pair that’s already vetted.

— Sachin Rekhi

It still is AI slop because we could never ship this. This would never be considered high-craft work.

— Sachin Rekhi

There’s actually 15 unique skills you kind of have to master to be able to do AI prototyping well.

— Sachin Rekhi

We should be using [AI] to create multiple outputs… a designer would come up with three variants.

— Sachin Rekhi

If a PM is trying to get a full version of their product out through these vibe coding tools, they’re doing it wrong.

— Sachin Rekhi

Product shaping vs traditional roadmappingAI slop: causes and avoidanceAI Prototyping Mastery Ladder (15 skills)Baselining via screenshot recreationBatching edits and versioning tradeoffsDiverging across tools for more variantsFunctional prototyping with real APIs (OpenAI)Secure handling of API keys/secretsCustomer validation at scale (surveys, analytics, session replay)Workflows vs agents (conceptual decisioning)Prototypes vs PRDs (discovery vs strategy)AI prototyping tools market map and picks

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome