Advanced Guide to AI Prototyping with Sachin Rekhi (Reforge)

Check out the conversation on Apple, Spotify and YouTube.

Brought to you by - Reforge:

Get 1 month free of Reforge Build (the AI prototyping tool built for PMs) with code BUILD

---

Today’s Episode

When you first start using AI prototyping tools, you get wowed.

You type “create me a CRM application” and boom a fully functioning app appears in 60 seconds.

But here’s the problem.

It looks generic. The styling is basic. The features are vanilla. You’d never ship this to customers.

This is AI slop.

Sachin Rekhi was the former Head of Product at LinkedIn Sales Navigator. He’s now teaching thousands of PMs at Reforge how to master AI prototyping.

And in today’s episode, he breaks down everything.

---

If you want access to my AI tool stack - Dovetail, Arize, Linear, Descript, Reforge Build, DeepSky, Relay.app, Magic Patterns, and Mobbin - for free, grab Aakash’s bundle.

Are you searching for a PM job? Join me + 29 others for an intensive 12-week experience to master getting a PM job. Only 9 seats left.

---

Newsletter Deep Dive

For paid subscribers, I’ve written up a complete guide to AI prototyping (that summarizes + goes beyond the pod). It’s the Ultimate AI Prototyping Masterclass:

1. Why AI Prototyping Changes Everything

2. The AI Prototyping Mastery Ladder

3. Design Consistency

4. Advanced Prototyping

5. The AI Prototyping Tools Face-Off

---

1. Why AI Prototyping Changes Everything

Most companies can’t do product shaping.

They prioritize problems first. Then they figure out solutions.

Here’s the typical flow:

You identify a customer problem → Put it on the roadmap → Start ideating solutions → Design mockups → Build in production

The issue is you’re committing to the problem before you’ve validated any solution actually works.

Anthropic does this differently.

They build prototypes for every customer problem they’re considering. They launch them internally. They see what people actually use. Then they decide what to productionize.

This is product shaping.

They’re prioritizing problem-solution pairs that are already validated. Not just problems.

This Used to Be Impossible

In the past, only companies like Apple could afford this approach.

Jony Ive ran a massive lab where he prototyped everything. He’d build dozens of versions of products, show them to Steve Jobs, then decide what to ship.

The iPhone itself came from this process. Ive had built a capacitive touch interface for a tablet. Jobs saw it and said “this would be better on a phone.” They canned the tablet project and created the iPhone.

But most companies couldn’t justify building expensive prototypes they’d throw away.

AI prototyping changes this.

You can now build sophisticated prototypes for essentially free. No massive lab required. No thousands of dollars spent on throwaway projects.

The product shaping approach previously only available to elite companies is now available to everyone.

The Three Reasons This Matters

Reason 1 - Speed

Traditional prototyping with Figma takes days or weeks. You’re creating static mockups. No interactions. No functionality.

AI prototyping takes minutes. And you get a working application.

Reason 2 - Functionality

Static mockups can’t capture real interactions. Users can’t click through flows. They can’t experience the actual product.

AI prototypes are functional. Users can interact with real features. You can integrate APIs. You can test with real data.

This means better validation. You’re not asking “does this design look good?” You’re asking “does this feature solve the problem?”

Reason 3 - Cost

Building prototypes in code used to require engineering time. Engineers would build throwaway projects that never shipped.

Now PMs can build prototypes themselves. No engineering time required. No expensive developer hours wasted on experiments.

The math is simple: more experiments, faster validation, better products.

But There’s a Problem

Most AI prototypes are slop.

When you type “create me a CRM” into an AI tool, you get something that looks like this:

1. Generic styling that looks like a wireframe

2. Basic features that exist in every other CRM

3. Simple scenarios with no product differentiation

This is magical from a technical standpoint. But you’d never ship this.

The challenge is going from AI slop to production-grade prototypes that actually represent your vision.

That’s what the rest of this guide teaches you.