When folks ask me how I received began designing with AI, I give them the sincere reply: I didn’t have a grand plan. I wasn’t getting down to reinvent my design course of or turn out to be an knowledgeable in a brand new software. I simply needed to see if I may breathe life into an previous sport concept that I had placed on the shelf. That curiosity led to experiments that modified the best way I take into consideration design.
With AI, designers are getting “nearer to the steel” than ever earlier than. Again within the day, being near the steel meant manually setting the kind for each letter on the web page. In software program design, we’ve usually been a step eliminated, handing off our visions to builders who translate them into code. Now, AI instruments are shrinking that hole. Whereas this definitely doesn’t exchange our engineering companions, designers are gaining a brand new form of inventive management.
You don’t want all of the solutions to get began with AI. What issues is curiosity, a real willingness to strive new issues, and the openness to be taught as you go.
Right here’s what I’ve realized in my journey:
Curiosity leads the best way
Techniques considering makes higher designers
AI is a real inventive collaborator
Roles are mixing and that’s an excellent factor
Phrases matter greater than ever
Shaping the longer term collectively
Curiosity leads the best way
I’ve all the time been the kind to poke at attention-grabbing new instruments simply to see what occurs. In experimenting with Cursor, I puzzled if it may assist me prototype an previous sport thought I’d had. I didn’t know every part once I began, and that was the purpose. Designers who thrive with AI don’t wait till they’re “prepared.” They’re those who discover, take a look at, and be taught — even when it means breaking a number of issues alongside the best way.
Generally my curiosity takes the type of skepticism. I’m enthusiastically on the lookout for causes to name BS on AI. Proper now, these instruments signify the typical of human intelligence on the web, and let’s be sincere, that common isn’t all the time inspiring. By pushing previous the typical with sharper questions and bolder experiments, I usually discover good moments hiding within the outcomes.
Can you employ AI to create this form utilizing solely prompts?
Considered one of my favourite workout routines is one thing I name “use your phrases.” I’ll problem myself or my workforce to recreate a form or a visible structure utilizing nothing however prompts. The purpose isn’t whether or not the AI nails it. It’s about stretching how we body concepts in language and seeing what we be taught within the course of. As a rule, the surprises spark new instructions I in all probability wouldn’t have discovered alone.
Techniques considering makes higher designers
I can write some code, however I’m not a coder. I’ve tried to be taught to talk pc extra instances than I can rely, however it’s by no means fairly caught. Right here’s the wonderful shift: With AI, the pc realized to talk human.
That doesn’t imply I don’t care in regards to the technical facet. In actual fact, AI has pushed me to assume extra in techniques. I don’t have to know each line of syntax, however I do want to grasp the ideas that form how a product works.
For instance, once I requested the coding agent to clarify state machines, the response was manner too technical. So I stated, “Clarify it like I’m 10.” The agent returned an important metaphor about how a merchandising machine doesn’t distribute sweet bars except an individual places cash into it. Then, I received it. This trade created a brand new psychological mannequin for me. As an alternative of designing for remoted screens, I’m now designing how a system strikes between states, like logged in or logged out, or gentle mode or darkish mode.
Again to the highest
Studying a bit about deployment helped me perceive what it actually takes to get a product reside — together with model management, publishing updates, and managing workflows. With the ability to body my design concepts past surfaces and within the context of a system is what resonates with builders.
Moreover, understanding the economics of those instruments improves your savvy. I spotted that if I typed “begin the server” into Cursor, it will dutifully spin issues up. And, it value me tokens. By studying a easy terminal command, I may have saved these credit.
AI is a real inventive collaborator
My a-ha second got here once I constructed a working prototype of a childhood sport my son created to move the time throughout automotive rides. Known as Soar Astronaut Soar, my son would run his fingers throughout the window and faux to leap over buildings and bushes. I didn’t want to jot down code. I described what I needed in plain language, let the agent do the heavy lifting, after which spent my time the place it actually mattered, which was tweaking the timing of the astronaut’s bounce till it felt good.
In an experiment with coding brokers, I constructed a easy sport by having a plain language dialog with the AI. There was no want to jot down code.
That is the place we get into vibe coding — or what I describe as “chatting with a bot to construct stuff.” It’s meant for this kind of situation whenever you wish to discover what’s doable. As an alternative of ready weeks for an engineer to implement a design, I can sketch and iterate on a prototype in actual time with AI.
Utilizing provocatypes in my Salesforce work
At Salesforce, I’ve use AI to create prototypes (state machines) or merchandise for nearly all of my tasks.
I constructed a completely purposeful voice‑to‑textual content chat prototype in two hours. It had gentle and darkish mode that shifted mechanically, ran on cell, and labored whenever you tapped the microphone. My product administration companions and I may take a look at an actual expertise versus debate a hypothetical buried in a slide deck. This “provocatype” — a provocative prototype — sparked a extra productive dialog in a single assembly than weeks of static wireframes ever had.
With the account agent that’s part of Gross sales Administration, I used AI to create an early model of the static design. Particularly, I needed to check the interplay between the badge, popover, and facet panel parts. Whereas we now have requirements for a way every element ought to be used and behave independently, we would have liked to see how they’d work collectively and tackle any points within the end-user expertise.
These moments remind me that AI doesn’t simply take directions. It collaborates. Once I describe my intent clearly, it helps form prospects I couldn’t attain alone. And when the prototype is in entrance of my workforce, we’re reacting to one thing tangible, not summary.
Again to the highest
Roles are mixing and that’s an excellent factor
Design, improvement, and product are nonetheless distinct crafts. However increasingly more, the boundaries between them are fluid.
I’ve shipped working apps as a designer, as an alternative of simply handing off specs. And I’ve additionally watched builders experiment with interplay design in instruments like Cursor, which flips the script in the easiest way. It’s not about doing one another’s jobs. It’s about increasing the area the place we meet, so we are able to ask the proper questions, transfer sooner, and make higher selections.
That overlap issues. It creates a shared language with which we are able to form the product collectively from the within out. In apply, which may imply I spin up a provocatype to check a function earlier than we even think about committing to a dash, or a PM drops a transcript into an AI software and shares a working draft app on the finish of a gathering. Once we every have the flexibility to prototype, take a look at, and iterate in actual time, collaboration stops feeling like a relay race. It begins feeling like co‑creation.
Phrases matter greater than ever
If you happen to can discuss, you may design with AI. I imply that actually.
These instruments perceive pure language now. The clearer I’m within the logic and intent of my prompts, the higher the outcomes I get. Readability doesn’t all the time imply polished, nevertheless. I’ve despatched prompts with typos, slang, and even the occasional curse phrase. And the AI nonetheless understood. That taught me that what issues most is speaking intent.
Experimentation can even embody emotion. You may convey urgency or curiosity with the phrases you employ.
I’ve additionally skilled the facility of phrases in workforce design settings. As soon as, after a protracted assembly stuffed with concepts, I merely pasted the transcript into an AI with the immediate: “Construct the app we simply mentioned.” Inside minutes, we had a working prototype to react to. The temper shifted immediately and the extent of engagement skyrocketed as a result of we may actively check out concepts.
Shaping the longer term collectively
What started as a small experiment has grown into a complete new manner of working as a result of AI is increasing the craft of design in methods I by no means imagined. For the primary time in years, I really feel like I can form not simply the imaginative and prescient of a product however the precise product itself. That doesn’t reduce the significance of engineers. Quite, we are able to collaborate extra deeply and remedy extra issues.
Finally, it’s not about mastering a selected software. Once we can do issues in minutes that used to take days, that permits us to focus our consideration on extra essential issues. Designing with AI can have an effect that goes past our identification as designers.
Again to the highest
