I still remember the first time I saw an AI-generated image. It was a few years ago, and a friend showed me a completely bizarre, yet oddly compelling, picture created by some early generative AI. My first thought was, "Okay, this is cool, but how is this actually going to change anything?" Fast forward to today, and AI has completely blown past what I thought was possible, especially for us in product management. What used to take weeks or even months to prototype, we can now do in a matter of minutes. Frankly, it's a game-changer.

This post will transform how you build products, come up with new ideas, and operate as a PM. Whether you’re already deep into AI tools or just getting started, you’ll learn what tools you should be paying attention to, which tool to use when, and how to get unstuck when you run into an issue. You’ll find a collection of battle-tested prompts, real-world examples, and a step-by-step guide you can put into practice immediately. Imagine being able to turn Figma designs into a working app with a few clicks, or turn your PRD into an interactive prototype in minutes. This is all possible, and you’ll learn how.

The AI Shift for Product Managers

If you haven’t been paying close attention over the past six months, you may have missed the rise of tools like Cursor, Replit Agent, v0, Bolt, and other new AI tools that allow you to build working apps in minutes. For example, it took me 10 minutes to build this 2-D tank game (with an AI opponent included), merely using this series of prompts:

“Build a 2d tank game with an AI opponent.”

“Add collision for the shot when it hits a tank.”

“When health hits zero, play an animation and reset the game.”

“Improve the acceleration for player movement.”

“Make it so holding down the space bar has a timer to shoot a 2nd time.”

“Add power ups to the map.”

Pretty cool. But what’s cooler is that you can use these tools to build functional prototypes from a Figma design, convert a rough hand-drawn sketch to a working app, translate a PRD document into an interactive prototype, or even build a usable internal tool for your team, with no coding ability. In this post, I’ll cover the basics of AI prototyping, show how to get good results out of the most popular tools, and walk through an end-to-end example of building a prototype in less than 10 minutes.

Choosing Your AI Tools

Current AI development tools come in three types:

• Chatbots (e.g. Claude, ChatGPT): These are the AI tools you probably already know, capable of writing and explaining basic code.

• Cloud development environments (e.g. Replit, Bolt, v0, Lovable): These are full-stack platforms that can build and run your apps in the cloud.

• Local developer assistants (e.g. GitHub Copilot, Cursor, Windsurf, Zed): These are development environments (IDEs) that help you write code with AI.

Let’s review the most popular tools in each category to see what they can do and what we can build.

Chatbots: Quick Code Generation

Best for: Simple, single-page prototypes without complex design needs, like calculators, flip cards, or data visualizations.

Chatbots can write code directly from your questions or prompts. For instance, a prompt like “Build me a calculator with React” will give you the code.

With ChatGPT, you'll need to copy and paste that code into your IDE and run it on your own computer. Claude goes a step further with its Artifact system, letting you run the code right inside its interface and deploy it to a shareable link. The catch is you can’t directly edit the code; you're entirely reliant on prompts for any changes.

Remember, chatbots can write code for any part of your stack (client, server, database) but won’t host or deploy it for you. They’re not great for complex, multi-page prototypes, and direct code changes are tricky. Use chatbots for very simple, one-off prototypes—a basic landing page, an individual input like a date picker, or a small to-do list app.

Cloud Development Environments: Building Full Apps

Best for: Prototypes with multiple features, specific design requirements, or many pages.

Cloud development environments are a big upgrade from chatbots. These tools handle nearly everything needed to turn your ideas into a working product. They can help you build end-to-end features, manage the backend, allow multi-file edits with agentic workflows, and handle more complex tasks, like updating your database schema.

A key difference among these environments is how they handle hosting. Every software product has three parts: a client (what the user sees, usually JavaScript), a server (processes requests, often Node.js, Python, or Java), and a database (for permanent data storage). Building prototypes with real features means hosting both your client and server code, and possibly a database.

Popular Cloud Development Tools

• v0: This tool can write and host both client and server code, defaulting to Next.js and Shadcn UI. It’s great for deployment and has excellent default styling. Here’s a basic CRM built in v0 with the prompt “Build me a basic CRM.”

• Bolt: Similar to v0, Bolt generates and deploys client and server code. However, Bolt runs server code directly in the user’s browser. This means it can’t natively support prototypes needing user logins, multi-user interactions, secure data operations, or persistent data storage. You can get around this by integrating with external products like Supabase.

• Replit: Replit allows you to build full-stack applications with client, server, and database. It works well with both JavaScript and Python and shines for internal admin tools (like file converters or job trackers) and data-driven apps (like image resizers or multi-page dashboards).

• Lovable: The newest of the bunch, Lovable is similar to v0 and Bolt, generating websites using JavaScript frameworks like React and Next.js. Its strength lies in integrations with tools like GitHub, Supabase (for authentication and databases), and AI providers like Anthropic and OpenAI. This makes it ideal for building production-ready apps.

To recap:

• Choose v0 for beautiful designs out of the box.

• Choose Bolt for quick prototypes with flexible designs.

• Choose Replit for internal tools or products that store or transform data.

• Choose Lovable for production apps that benefit from integrations.

All cloud development environments allow you to build more complex applications than chatbots, deploy to the cloud, and easily share updated versions.

Local Developer Assistants: Coding with AI Support

Best for: Product managers who have some coding experience and are working on serious applications for production.

Local developer assistants are aimed at people who know how to write code. Tools like Cursor and GitHub Copilot can take prompts similar to Claude but then generate and apply changes directly within your codebase and development environment (IDE). They go beyond autocomplete, now writing most of your code from prompts alone.

For example, I built a presentation app (with live Q&A and polls!) in about 10 days using Lovable and Cursor. I started in Lovable for basic features, synced to GitHub for editing in other tools, and used Cursor for final changes and bug fixes. This app uses authentication, databases, real-time updates, and more.

Ten days might sound like a lot, but most of that time was spent resolving bugs and troubleshooting—something Cursor excels at.

GitHub Copilot is popular in enterprise settings due to its trusted vendor, Microsoft. It supports multi-file changes from prompts and code explanations. I've found it works best with very specific directions; it doesn't perform as well as Cursor with general instructions.

Windsurf and Zed are two other IDEs with AI capabilities. Windsurf suggests multi-line changes and file operations, excelling with larger codebases. Zed is a high-performance editor with features like prompt libraries and AI-generated code application.

Building Your AI-Powered Prototype

Now that we’ve covered the basic tools, let’s look at two common prototyping scenarios for product managers:

• Converting an existing design into a functional prototype.

• Building a prototype from scratch based on an idea.

Scenario 1: Converting a Design

Let’s say you want to turn a design for a popular home-sharing platform into a working prototype to explore a new feature, like a price filter.

[Insert initial design screenshot here]

I chose Bolt for this because it’s great for building from existing designs, and we don’t need a backend database. Here’s a prompt you can use (and a pro tip: be hyperspecific with subsequent prompts to help the AI pinpoint changes):

Prompt: "Build a prototype to match this design. Match it exactly. Use Tailwindcss.

Match styles, fonts, spacing, and colors.

[Include a single screenshot of the design]."

After that, we can add our new price filter feature:

Prompt: "Implement an inline price filter as a component of the search bar. It should appear next to ‘Add guests’ in its own section. Selecting the input should pop up a price filter with minimum and maximum values. The background of the pop-up should be white and should cover elements beneath it."

This gives us a great starting point. Let’s add a slider for the minimum price:

Prompt: "Can you add a price slider? It should have a blue line and a black node. Sliding the node should modify the minimum price."

In just 10 minutes, you have a functional prototype of your product idea, all without writing a single line of code! You could even keep refining it, showing real-time listing updates as you adjust the price. Check out the prototype here.

Scenario 2: Building from Scratch

If, like me, your design skills aren’t top-notch, you can still build prototypes using existing patterns from free design systems like Tailwind or Shadcn UI.

Let’s build a quick CRM with Bolt and then add a new feature. Imagine you’re considering an automated email outreach feature and want to get customer feedback with a prototype.

First, a basic CRM:

Prompt: "Create a comprehensive customer relationship management (CRM) system."

We just created a working CRM prototype in under five minutes—something that used to take weeks of engineering time! Now, let’s add that new feature:

Prompt: "Please implement a mock AI email writer. This should be accessible from the left nav."

Again, less than five minutes to explore a new feature idea. Think about how many ideas you can test and how quickly you can bring them to market. AI prototyping tools have completely changed how fast teams can ship. You can now get interactive examples into your customers’ hands much earlier, speeding up your discovery process.

Each AI tool will give different results based on its default settings and how specific your prompts are. Here’s the same example with v0, using the same prompts.

Essential Prompt Templates for PMs

Here are some templates to help you get started with the tools we’ve discussed:

Task #1: Build a prototype from an existing Figma design

Tool: Bolt

Prompt:

Build a prototype to match this design. Match it exactly. Use Tailwindcss.

Match styles, fonts, spacing, and colors.

[Include a single screenshot from Figma]

Example: Deployment manager

Related Articles

• The Future of Product Managers is Here: Meet Your AI Co-Pilot

• The Next Evolution in Product Tooling: How AI Changes Everything

• Product Managers, Reinvented: Embracing Your Role as AI Orchestrators