Bargaining Chip

Inspiration:

I wanted to build something that served a social good—and while weddings may not seem like a typical cause, the financial pressure they create on families is very real. Prices for venues, caterers, and vendors vary wildly, and there’s no centralized place where people can check if they’re being overcharged.

So I created Bargaining Chip: a space where couples can anonymously share what they paid, what they negotiated, and what they learned. It helps others save money while also keeping the wedding market in check, which today runs more on perception and negotiation than fairness or transparency.

What it does:

Bargaining Chip is a community-driven platform where users can:

Share their wedding experience via a friendly AI chatbot
Submit pricing and vendor details (venue, caterer, DJ, etc.)
Search and explore reviews based on location, budget, and wedding style
Match with similar past weddings based on their own wedding goals and budget
Use real vendor pricing as a reference point while planning their own wedding

This creates a database of real wedding costs, negotiation insights, and vendor ratings to help future couples plan smarter.

How we built it:

We built Bargaining Chip primarily using Bolt.new for the frontend experience and UI flows. For the backend, we used Supabase to manage user data, structured reviews, and matching logic.

To collect detailed wedding experiences, we designed a conversational review system using Gemini GPT, which acts like a friendly bot that talks to married users and extracts structured data around vendors, pricing, and experiences.

Key components:

Bolt.new: UI, user journeys, search experience
Supabase: Backend for review storage and matchmaking logic
Gemini GPT: AI bot for collecting rich, conversational user data and converting it into structured summaries
Matching logic: A lightweight rule-based algorithm for finding relevant wedding stories based on budget, guest count, and location

Challenges we ran into:

Bolt AI inconsistencies: The GPT-based chatbot built within Bolt would often malfunction, especially when we tried training it for structured data extraction. Fixes were temporary, and the same issues sometimes resurfaced, making it hard to rely on.
Chatbot complexity: Training the bot to ask natural follow-ups and still collect all necessary vendor and pricing details took almost a full day, especially since I was doing it for the first time.
UI responsiveness issues: Some layout bugs would reappear unexpectedly, and I had to rebuild sections of the app to maintain consistent UX.

Accomplishments that we're proud of:

Successfully built a functioning conversational review system in Bolt that feels friendly and intuitive
Designed a matchmaking algorithm that connects users with similar wedding profiles
Created a clean, informative search interface for vendor pricing
Managed to keep the experience localized and globally scalable

What we learned:

Bolt is powerful but has limitations, especially when working with AI components that need precision and memory.
Building a conversational UI is more than just good design—it requires flow logic, fallback options, and contextual clarity.
Even in a hackathon, getting structured data from unstructured stories requires thoughtful prompt engineering and testing.
And most importantly, community-powered transparency is a product design challenge—but also an exciting opportunity.

What's next for Bargaining Chip:

Invite vendors to host their businesses on our platform, so users can directly search for them, view real reviews, and even make a demo booking
Highlight and promote honest, fair vendors who align with our mission of transparency and trust
Maintain our focus on being a community-driven utility, not just a commercial platform
Continue improving the matchmaking system to offer smarter recommendations
Explore adding localized pricing trends and alerts for overpriced deals
Expand to other countries where wedding pricing is also opaque and bargaining-based