From Scribble to Product: How I Think with an AI Whiteboard

For the longest time, my relationship with artificial intelligence was purely textual. I’m an editor. My world is, and always has been, built on words. I live in documents, I communicate in long emails, and my primary creative output is a string of carefully chosen characters. So when tools like ChatGPT arrived, they fit neatly into my existing world. AI was a conversational partner, a tireless writing assistant, a machine you fed text to and got text from. It was powerful, yes, but it was also an echo chamber for my own text-based thinking.

My desk in our Mumbai office is a testament to this limitation. It’s perpetually cluttered with notebooks filled with frantic scribbles, mind maps that trail off into nothingness, and diagrams that only I could ever decipher. These were the ghosts of ideas, visual thoughts that were too clunky to translate into a formal document and too messy to share. They were born on paper and, for the most part, they died there. The friction between a visual idea and a shareable digital artifact was just too high.

Then, a few months ago, I stumbled across a video that fundamentally rewired my understanding of what AI could be. It was a demonstration of a tool called tldraw, specifically its ‘Make Real’ feature. Someone drew a crude, wobbly sketch of a user interface on a digital whiteboard, hit a button, and the AI transformed it into a clean, functional-looking mockup, complete with code. I was floored. This wasn't text-in, text-out. This was scribble-in, reality-out.

My initial skepticism gave way to intense curiosity. An AI that could interpret the chaotic language of my brainstorming sketches? It seemed too good to be true. I immediately opened a blank canvas and decided to give it a try. I drew a very basic flowchart for our editorial workflow at Pulse AI. It was nothing more than three lopsided boxes with the words 'Draft', 'Review', and 'Publish' connected by jagged, uncertain arrows. It was the kind of drawing you’d make on a restaurant napkin.

I selected my clumsy creation, held my breath, and clicked 'Make Real'. The transformation was instantaneous. My lopsided boxes snapped into perfectly aligned rectangles with clean fonts. My wobbly arrows became crisp, sharp connectors. It was… beautiful. It was my idea, but polished and legible. In that single moment, a new mental model clicked into place. AI wasn't just a language processor; it could be a visual interpreter. It could understand intent, not just words.

This discovery has since branched into three major areas of my work, completely changing how I develop and communicate ideas. The first, and most straightforward, is using it as a high-speed diagramming tool. Before, if I needed to create a system diagram or a content strategy map, it was a whole production. I’d have to stop thinking, open a dedicated app like Figma or PowerPoint, and then meticulously drag, drop, and align shapes. It was a chore that broke my creative flow. The tool got in the way of the thought.

Now, I just stay on the infinite canvas of the whiteboard. I can sketch out the architecture for a new feature we're considering, mapping how a reader in Bangalore interacts with our app, how data flows to our recommendation engine, and how we surface a new content summary. The process is fluid and fast. I draw, I write, I connect. When I’ve dumped all my thoughts onto the page and it looks like a complete mess, I simply select sections and let the AI tidy them up. It’s like having a junior designer on call, ready to turn my frantic sketches into professional-looking artifacts in a split second. It removes the friction, allowing me to stay in a state of pure ideation for longer.

Second, it has revolutionized how I brainstorm. Traditional mind mapping, even with digital tools, can feel surprisingly linear. You start from a central point and branch out. It’s organized, but not always how truly creative, chaotic thought happens. The AI whiteboard, however, encourages a more spatial and unstructured approach. I can throw anything onto the canvas. I can draw a crude lightbulb next to a headline idea. I can sketch a quick Venn diagram to compare two different market entry strategies for a new product. I can draw a simple timeline for a project launch.

The AI acts as a creative accelerant. I don't have to search for an icon pack; I just draw a lightbulb and 'Make Real' turns it into a clean graphic. I don't have to fiddle with drawing tools; I sketch a Venn diagram and the AI renders perfect, overlapping circles. This isn't just about aesthetics. By making it effortless to introduce visual elements, the tool encourages me to think more visually. The brainstorming sessions, even when I’m alone, feel more dynamic. The canvas becomes a living document, a playground for ideas rather than a rigid structure for them.

Perhaps the most impactful change, however, has been in creating simple UI and UX mockups. As an editor, I’m deeply invested in how our readers experience our content. I often have ideas for improving our website or app, but communicating these to our product and design teams in Pune used to be a struggle. I'd write a long, descriptive document trying to explain a new layout, or I’d wave my hands around in a video call. It was imprecise and often led to misunderstandings.

Now, I just draw it. If I have an idea for a new homepage layout, I'll open the whiteboard and sketch it out. A large rectangle for the lead story, three smaller boxes below for featured articles, a thin rectangle on the side for a newsletter signup form. It's the crudest of sketches. But then I select it all, hit the magic button, and the AI generates a clean, simple wireframe. It will even generate the HTML and Tailwind CSS code for it. It’s not a final, pixel-perfect design, and it’s not meant to be. Our brilliant designers handle that.

But what it is, is a high-fidelity conversation starter. It's a tangible prototype. I can share that link with a developer or designer and they instantly get what I’m thinking. The conversation is no longer 'So, Rohan is imagining a box here…' but 'Okay, I see the layout. What if we moved this component over here?' It has democratized the early stages of product design. I'm no longer just 'the words guy' trying to describe a picture; I can actually provide the picture. It has made my collaboration with the product team ten times more effective.

This shift from a text-based AI to a visual AI partner represents something profound. It’s not just about making pretty diagrams or faster mockups. It's about externalizing thought. It closes the gap between the fleeting, messy ideas in your head and a concrete, shareable object embodying that idea. It’s a form of ‘thinking in public’ at the earliest stage, making your creative process visible and malleable.

For anyone in a creative or product-focused role who feels like their best ideas are trapped in notebooks or hard-to-explain thoughts, I cannot recommend this new class of tools enough. It's not about learning to code or becoming a designer. It's about finding a more intuitive, more human way to communicate with and through technology. AI is finally breaking free from the chatbox, and instead of just helping me write, it’s helping me see.