Simplified AI Bot Builder - So easy your grandma can use it.

Client

Botpress Lite

Year

2025

Most AI chatbot builders are made for developers, not everyday users. Even “no-code” tools often bury people under unfamiliar terms, logic flows, and overwhelming interfaces.

Botpress Lite posed this challenge:

“Design a chatbot creation experience so simple that anyone, even someone with no technical background, can build a working AI assistant in minutes, right from their phone.”


Scope of Work

Replit Development
Rapid Prototyping
Wireframes
Functional Code
Vibe Coding

Project Goals

The goal wasn’t just to make a chatbot builder, it was to make it feel effortless and friendly.

I focused on creating an experience that was:

  • Approachable: No jargon, no coding, no setup anxiety

  • Fast: Just a few taps to get something working

  • Conversational: Natural language first, forms second

  • Mobile-first: Prioritized for small screens and thumbs


Understanding the User

Since this was a solo design challenge with no access to real users, I built a set of light personas based on common patterns seen in users of tools like Wix, Notion AI, and Canva:

  • Sofia (32) — small business owner who wants a support bot on her website but doesn’t want to hire a developer.

  • Tony (45) — works in HR and wants an internal vacation request bot but has never built a chatbot before.

  • Emily (58) — runs a nonprofit and just wants an “ask me anything” bot for volunteers.

These users:

  • Know what outcome they want

  • Don’t want to learn how bots or flows work

  • Feel overwhelmed by traditional builder UIs

This shaped my core design principle:

Focus on outcomes, not options.

Design Strategy

1. Reduce Cognitive Load

I used a stepper to divide the process into bite-sized, focused tasks. Each screen has one goal, one CTA, and only the information the user needs right now.

This taps into:

  • Progressive Disclosure — reducing overwhelm

  • Goal Gradient Effect — motivating completion by showing visible progress

  • Jakob’s Law — making navigation predictable

2. AI-First Input

Instead of starting with forms or toggles, the experience begins with a simple text box:

“Tell us what kind of bot you want to make.”

Users can describe their bot in plain language:

  • “A bot to answer questions about our menu”

  • “A customer service chatbot for a Shopify store”

  • “A bot that helps with internal IT tickets”

From there, the system applies smart defaults to build the rest of the experience.

3. Abstraction of Complexity

Behind the scenes, every bot includes essential features:

  • Pre-built fallback flows

  • Common intents like greetings or handoff

  • Smart suggestions for improvements

But the user never sees logic trees or entities, unless they ask to.

This keeps the focus on intent, not configuration.

4. Mobile-First Design

Most existing builders are desktop-centric. I flipped the assumption and designed for mobile from the start:

  • Full-width buttons

  • One action per screen

  • Large input fields and generous padding

  • Consistent iconography and visual hierarchy

The UI is inspired by Material Design and Apple’s Human Interface Guidelines to ensure accessibility and consistency.

Prototype & Flow

I built a working prototype in Replit that walks through the entire bot creation journey on mobile.

Screens include:

  • Welcome and AI prompt

  • Auto-generated bot preview

  • Optional customization

  • Confirmation and test mode

Watch below:


What I Learned

This challenge wasn’t just about UI polish — it pushed me to think about how non-technical users relate to AI. Here’s what I learned:

  • Clarity > Power: Smart defaults and limited choices create trust, not frustration.

  • UX Writing Matters: Good microcopy makes users feel supported, not lost.

  • Trust is Earned: Preview screens, undo actions, and confirmation steps matter when users are trying something new.

  • AI Isn’t Magic Without Framing: Users need to know what they can ask the AI to do — otherwise, it feels like a blank page.

What I’d Explore Next

If this were a live product, I’d iterate on:

  • A short onboarding wizard to teach the AI prompt format

  • Visual bot testing where users can interact before publishing

  • An “expert mode” toggle to show advanced tools when needed

  • Usage-based suggestions: “Most bots like yours include an FAQ flow — want to add one?”


Final Thoughts

Designing for simplicity is hard. It means constantly asking:

“What can we remove without losing clarity?”

This project helped me sharpen my product thinking, test assumptions about non-technical users, and craft a UX that gets out of the way, so users can focus on what they want to build.


Like what you see?
Book a free discovery call.

Like what you see?
Book a free discovery call.