How I Use AI: Think Browser-First Design, Not Figma Perfection

If you’ve ever designed directly in the browser instead of starting with mockups in Figma or Sketch, you’ll immediately understand how I approach AI tools. It’s the same mindset: quick, iterative, and focused on getting from idea to working prototype as fast as possible.

The Browser Design Parallel

When I design in the browser, I’m not worried about pixel-perfect layouts or choosing the perfect shade of blue. I’m throwing together HTML and CSS to see if an idea works. The code is messy, the styling is rough, but I can test the concept immediately. Does this navigation pattern actually make sense? Is this layout readable? Can users accomplish what they need to do?

This is exactly how I use AI.

AI as My Rapid Prototyping Engine

The Quick and Dirty Phase

Just like browser-first design, my initial AI interactions are exploratory and rough around the edges. I’m not crafting perfect prompts or expecting polished output. Instead, I’m:

• Testing ideas quickly: “Give me three different approaches to this problem”
• Exploring possibilities: “What if we approached this from a completely different angle?”
• Getting unstuck: “I’m stuck on X, help me brainstorm some alternatives”
• Generating raw material: “Create a rough outline for this concept”

The output isn’t publication-ready, just like my initial browser sketches aren’t production-ready. But it gives me something tangible to react to, critique, and build upon.

Why This Approach Works

Speed over perfection: I can test 5-10 different directions in the time it would take to perfect one approach traditionally.

Lower stakes experimentation: Since I know I’ll refine later, I’m more willing to try wild ideas that might not work.

Immediate feedback loop: I can see results instantly and adjust course, rather than spending hours on something that might be fundamentally flawed.

Reduced creative paralysis: Having something imperfect to work with is better than staring at a blank page.

The Refinement Phase

Here’s where the browser design analogy really shines. After I’ve found a direction I like through rapid AI prototyping, I shift gears completely.

From Rough to Refined

Just as I’d take a promising browser sketch and rebuild it properly—with clean code, proper architecture, and attention to detail—I do the same with AI-generated content:

• Structure and organize: Take the scattered ideas and arrange them logically
• Add my voice: Inject my perspective, experience, and unique insights
• Fact-check and verify: Ensure accuracy and credibility
• Polish the language: Refine tone, flow, and clarity
• Test and iterate: Make sure it actually solves the original problem

The Human Layer

This refinement phase is where my expertise becomes crucial. AI helped me explore the possibility space quickly, but now I need to apply judgment, domain knowledge, and creative thinking to make it genuinely valuable.

It’s like the difference between a wireframe and a finished design, both are necessary, but they serve different purposes in the creative process.

When This Approach Works Best

This browser-first mentality with AI is particularly effective for:

• Content creation: Blog posts, documentation, marketing copy
• Problem-solving: Breaking down complex challenges into manageable parts
• Research and analysis: Gathering information and identifying patterns
• Creative brainstorming: Generating ideas for projects, campaigns, or solutions
• Learning new topics: Getting up to speed quickly on unfamiliar subjects

The Mindset Shift

The key insight is treating AI as a prototyping tool rather than a production tool. Just as you wouldn’t ship that quick-and-dirty browser sketch to users, you shouldn’t publish raw AI output without refinement.

But also just as browser-first design can lead to better final products by letting you test ideas quickly, AI-first ideation can lead to better outcomes by helping you explore more possibilities in less time.

Making It Work for You

To adopt this approach:

1. Lower your initial quality bar: Aim for “good enough to evaluate” not “ready to ship”
2. Embrace iteration: Plan for multiple rounds of refinement
3. Stay in control: Use AI to enhance your process, not replace your judgment
4. Know when to stop: Don’t over-rely on AI when human expertise is what’s needed

The goal isn’t to eliminate human creativity, it’s to amplify it by handling the heavy lifting of exploration and initial drafting, freeing you to focus on the higher-level thinking that only you can provide.

The Bottom Line

For me, AI works best when I treat it like designing in the browser: fast, experimental, and iterative. I use it to prototype ideas quickly, then apply my expertise to refine and perfect the results. It’s not about replacing the design process, it’s about making the early stages faster and more exploratory, so I can spend more time on what really matters: creating something genuinely valuable.