Long story short
In 2025 I led the design of UI Bakery’s AI Chat Agent - the release that turned a traditional low-code builder into a place where you build a full internal app from a single prompt. As the sole designer, I shaped every layer: the conversational flow, prompt-to-app generation, the motion-feedback system, data integration, and how the design system stretched to cover AI features. I also built ready-to-ship motion prototypes with Claude so the AI interactions felt alive instead of laggy. The result was a fast, trustworthy AI experience that pushed activation ~10× past the main platform.
Problem
Going from a structured builder to an open-ended, prompt-driven product, low-code still felt like building, not generating - the AI only assisted the workflow, and slow, shapeless feedback chipped away at trust and time-to-value.
Solution
As the sole designer, I designed the AI experience end to end - prompt-to-app generation, a motion-feedback system prototyped with Claude, flexible data integration, workflow and code export, and a templates-driven fast start from landing to builder.
Result
Activation jumped ~10× over the low-code product, with up to 15× faster app delivery, ~2-minute generation, and ~86% UX satisfaction - repositioning UI Bakery as an AI-driven platform.
Context
Once UI Bakery was a mature low-code platform, we started asking how far AI could push internal app creation. We moved step by step, and I designed several key AI releases along the way:
2023
UI Bakery AI Assistant
Chat-based helper inside the builder that supports debugging, explains code, and connects to documentation.
April 2024
UI Bakery AI App Generator (PH link)
The first version of the AI Agent that could generate internal tools and admin panels from SQL data.
October 2025
UI Bakery Agent for internal app generation
A reworked version of the agent that generates complete internal tools from a single prompt - with data connections, logic, and simple UI animations included.
Read more
The AI Chat Agent kicked off in mid-summer 2025 with a small crew - 4 engineers, 3 PMs, and me as the only designer.
The goal was to fuse conversational AI with the builder: describe what you want, and the system hands back a deployable app.
! Problem
As UI Bakery turned AI-first, the ground shifted - from a structured builder to an open-ended, prompt-driven experience. That brought a few real problems:
Low-code felt like building, not generating - users were pushed into developer-style thinking.
The AI assisted the workflow but didn’t replace it, so the efficiency gains stayed small.
Slow, unstructured feedback drained trust and dragged out time-to-value.
Design challenges
A few threads ran through the whole project:
Bringing AI to production. Turning AI from a demo into a real app builder - with data, logic, and clean UI.
Designing the AI feel. Using motion and micro-interactions to make the AI feel fast and clear.
System & ownership. Evolving the design system and owning the product end to end, solo.
Solutions
What was done and what benefits were achieved for each solution:
Prompt-to-App
A conversational interface - the industry standard now - where you just describe the app (“I need an order-management dashboard connected to my PostgreSQL data”).
The AI generates the full UI, structure, and logic in ~2 minutes, depending on the request.
Smart visual & motion feedback
Every step of generation was visualized with smooth motion cues, so the AI felt natural and alive instead of stuck.
I used Claude to craft purposeful animations and handed developers ready-to-use motion prototypes they could drop straight in - so there was nothing to adapt, and these small touches shipped much faster.
Workflow generation, code preview & export
The system builds triggers, actions, and conditions in the workflow editor automatically. Flip to Code view to review, edit, or export the generated logic and files.
We reused solutions we’d already built, which cut time-to-market.
Fast start & templates
I designed the flow from an animated landing page into an organization workspace stocked with ready-to-use templates and an AI input - type your first prompt, pick a starting point right after login, and you land straight in the builder, ready to connect data and generate.
Time-to-value dropped to minutes - people start building real apps within seconds.
Research, Discovery & Validation
AI Chat Agent was rolled out in early October 2025.
For validation, I used different techniques:
Customer feedback
Steady input from users through support and product analytics.
Behavioral analytics
PostHog to watch sessions and read how people actually behaved.
User testing
User tests, quick corridor checks, and customer interviews.
Impact & Results
“Wow! I tried building apps with UI Bakery AI - that’s really mind-blowing! Works so good… Impressive!”
UI Bakery enterprise client who previously used only the low-code platform.
Business outcomes
Higher activation in the AI Agent than the low-code product
Up to 15× faster app delivery
Better conversion and retention across the funnel
Strong lead-to-activation growth
Positioned the product as AI-driven
Exact numbers are under NDA; these are directional.
Design leadership impact
Led the AI Agent end to end - UX, flows, motion, feedback
Turned AI delays into smooth, predictable interactions
Scaled the design system to cover AI-generated UI
Held quality at speed (~72% SUS, confirmed by ongoing monitoring of users)
Reflections and Takeaways
I helped move UI Bakery from a theme tool to an AI platform where teams build apps from a single prompt - landing several times the activation of the low-code product. The AI Chat Agent was the turning point: a conversational tool that generates full apps in minutes. I designed the whole flow - first prompt, generation, feedback - keeping it clear, interactive, and trustworthy.
Reflection
This one proved that conversational UX isn’t really about chat - it’s about behavior, timing, and trust. Where the low-code builder needed long, predictable flows, here the job was to make a short interaction feel alive and confident. Designing with AI became about turning intent into real outcomes through clarity and responsive feedback.
