MarryMap

MarryMap: an AI wedding co-pilot that finds vendors, manages guests, automates outreach, and turns wedding planning into one organized shared space.

Comment

About MarryMap

MarryMap was inspired by a simple problem: wedding planning is exciting, but the work is fragmented. Couples manage vendors through WhatsApp, guest lists in spreadsheets, budgets in notes, event details in family chats, and inspiration across many websites. We created one intelligent workspace that brings this process together.

MarryMap is an AI-powered wedding co-pilot. Couples can ask the planner to research venues, photographers, court-marriage support, and other wedding services. Rather than returning plain text alone, MarryMap presents source-backed vendor cards with the original website, public profile image, map link, and publicly listed contact details when available. Users can save useful results to their wedding shortlist and continue planning.

The platform also includes:

  • Guest management and bulk guest upload
  • Tasks, budgets, timelines, and vendor shortlists
  • Wedding website templates with AI and manual editing
  • RSVP-ready publishing with a shareable link
  • WhatsApp-first communication, with email and calls as fallbacks

How We Built It

We built MarryMap as a full-stack TypeScript application. The frontend uses React and TanStack Start, while the backend provides AI research, data-enrichment, and workflow APIs. Supabase stores wedding, vendor, guest, and planning data. We integrated OpenWA for WhatsApp connection and QR-based device linking, then deployed the application on Google Cloud Run.

A central technical goal was trustworthy AI research. Vendor websites often contain incomplete or inconsistent information, so our enrichment flow extracts only publicly available details from cited source pages, such as contact details, profile images, and map links. If a source does not publish a detail, MarryMap says so rather than inventing it.

Challenges We Faced

The main challenges were handling inconsistent web data, designing a readable research interface, supporting WhatsApp sessions after deployment, and coordinating the web app, API, Supabase, source extraction, and Cloud Run infrastructure.

We also needed to turn AI responses into saved, actionable wedding records rather than temporary chat output.

What We Learned

We learned that an AI assistant is most valuable when it preserves context, shows sources, and helps people take the next action.

GPT-5.6 and Codex accelerated development by helping refine the research-card experience, source-enrichment logic, responsive UI, deployment configuration, and integration debugging.

MarryMap helps couples move from “Where do we start?” to an organized, collaborative, and shareable wedding plan.

Built With

Share this project:

Updates