Meet us live at LEAP 2026
Book a meeting
All Case Studies

— Case Study

AI & Machine LearningHospitality & Travel

AI Event Planner

An event planning startup needed a better way to run client consultations for couple-centered events. We built a stateful, LLM-powered conversational web interface that serves as the UI layer for a complex planning session, guiding couples through a multi-stage journey to generate a tailored couple brief.

AI Event Planner

— The Challenge

The Problem We Solved

The client needed a dynamic frontend application capable of handling a complex conversational AI workflow without losing user context or corrupting state.

  1. The frontend had to act purely as a UI layer for a stateful planning session, displaying suggestions but never modifying memory directly, relying entirely on backend API responses.

  2. Managing a strict 11-stage progression (S1 through S11), where the application had to handle dynamic backend decisions like advancing, staying, reanchoring, or requesting clarification.

  3. Processing complex conversation turns that combined custom user text, selected suggestion chips, and multiple image uploads (up to 10MB each) via multipart/form-data.

3

Challenges Identified

Every challenge was systematically addressed through tailored engineering and design — no workarounds, no compromises.

— Our Solution

How We Solved It

We built a frontend architecture optimized for real-time interaction with a versioned, canonical memory backend.

A responsive UI that adapts to the current conversational stage, transitioning from collecting basic logistics like location and date to curating personality tags and overall event vibes.

A unified messaging system that accurately stringifies selected UI chips alongside custom text, and validates and packages image uploads for backend AI vision analysis.

Dedicated UI components that render artifacts generated by the AI, such as auto-generated couple briefs and design direction cards, immediately after specific stages are completed.

— Tech Stack

Technologies Used

React
TypeScript
Tailwind CSS
Vite
OpenAI API
Vector Search

— Impact

Results & Outcomes

5 of 11
Planning Stages Live
Zero
Lost Context Between Sessions
Text + Photos
Combined in One Message

Couples never have to repeat themselves. Every answer is saved the moment it's given, even if they close the tab and come back the next day.

The moment a couple finishes describing their vibe, their personalized brief generates automatically. No extra step, no waiting on a planner to draft it by hand.

If the AI service fails or a response doesn't validate, the couple sees a clear retry message instead of a broken conversation or lost progress.

Couples can drop in a photo of a venue they love alongside their own words, instead of trying to describe a look and feel in text alone.

Start a Project

Want results like these?

Tell us about your project. We’ll respond within 24 hours with a concrete plan or an honest answer about what’s possible.

Explore more of our work across industries and technologies.

Chat with us