Let me take you back a little. In 2018, UX design was slow, labor-intensive, and often frustrating. I spent hours, sometimes days, on user research, gathering insights, creating wireframes, and building prototypes. Every iteration involved manual work, multiple approvals, and constant adjustments. Back then, artificial intelligence felt like a novelty: auto-layouts, basic color suggestions, and perhaps a logo generator that resembled clip art.
Fast forward to 2025, and the landscape is completely different. Today, I can create an interactive prototype in minutes, test hypotheses without launching a website, generate dozens of visual concepts, and predict user behavior before a single line of code is written. Tools like Figma AI, RunwayML, Framer AI Builder, Adobe Firefly, Midjourney, DALL·E, Attention Insight, Uizard, and Galileo have reshaped the entire UX workflow – from initial research to final product delivery.
Stage 1: Discovery & Research
How it was:
We conducted in-depth user interviews, surveys, and observations in real environments. Reports and analytics were collected manually, and understanding the user’s world often required physically immersing ourselves in it—tools: paper surveys, Excel, Google Forms, and handwritten notes.
How it is now:
AI can analyze user data almost instantly. I rely on Notion AI, ChatGPT, and Airtable AI to structure insights, create user personas, and detect behavioral patterns. Tools like Hotjar and FullStory let me see how people interact with interfaces in real time.
Stage 2: Define Problems
How it was:
After gathering data, we manually formulated user pains and created persona maps. Tools like Miro or Lucidchart were essential.
How it is now:
AI helps organize insights and identify key issues automatically. I use Figma AI for visual mapping, Notion AI for analytical reports, and Jasper AI for problem statements and user needs. The result: a clear, actionable UX strategy ready for the next stage.
Stage 3: Ideation
How it was:
Brainstorming sessions involved whiteboards, sticky notes, and Google Docs. Hours of discussion often produced ideas that were never implemented.
How it is now:
AI generates dozens of concepts in minutes: layouts, interaction patterns, interface elements. I use Figma AI for interface ideas, Miro AI for ideation maps, and RunwayML for visual concepts. This allows me to focus on solutions that actually solve user problems.
Stage 4: Wireframing
How it was:
Wireframes were created manually, screen by screen. Tools: Balsamiq, Sketch, Adobe XD.
How it is now:
Uizard and Galileo transform simple text prompts into editable wireframes. I write something like “Sign-up screen for a travel app in minimalist style” and get a ready-to-edit layout in seconds. Figma AI helps automatically arrange blocks and components, while Whimsical visualizes user flows.
Stage 5: Prototyping
How it was:
Interactive prototypes took hours: connecting screens, building component libraries, and scripting user actions. Tools: Figma, Proto.io, InVision.
How it is now:
Plugins like Figma AI, Framer AI Builder, and Adobe Firefly create interactive prototypes automatically buttons, transitions, copy, even basic animations. Prototypes come to life almost instantly, allowing me to test ideas immediately with my team or client.
Stage 6: Visual Design
How it was:
Visual concepts were created manually: searching for references, photo shoots, retouching, and rendering. Tools: Photoshop, Illustrator, Sketch.
How it is now:
Tools like Midjourney, DALL·E, Adobe Firefly, and RunwayML allow me to generate dozens of visual options in minutes. I can quickly present multiple concepts to clients and choose the best one. My role now is to guide AI and ensure visual style and quality, rather than do all the work by hand.
Stage 7: Testing
How it was:
Testing required real users, lab setups, A/B tests, and post-launch analysis. Tools: Hotjar, Google Analytics, UserTesting.
How it is now:
AI predicts user behavior even before coding starts. Attention Insight generates gaze heatmaps, while Maze and Lookback.io simulate user scenarios. I can identify issues and optimize interfaces without waiting for live testing.
Looking back at my journey as a UX designer, it’s clear: AI doesn’t replace me, it makes my work faster and more precise. Before, I spent days on repetitive tasks; now, I focus on strategy, logic, and product meaning.
Design no longer starts with a blank canvas. It starts with the question: “How will users interact with this product, and how can AI help me create solutions faster?”
AI accelerates the process, but the depth of thinking, taste, and responsibility remain mine. This is what transforms me from an operator of tools into a creator of next-generation interfaces.
