PriceMap

Because of new implementation of new dynamic pricing systems in major retailers like Walmart and Target. Customers need to check for them in webpages. PriceMap facilitates the search.

Comment

Inspiration

Miami currently faces one of the highest inflation rates in the country. With retailers moving toward AI-driven dynamic pricing, the "Miami Tax" changes by the hour depending on which neighborhood you're in. We wanted to give the power back to the consumer by creating a tool that treats grocery shopping like the stock market.

What it does

PriceMap (or your chosen name) is an AI-powered logistics agent that analyzes a user's grocery list against a dynamic local database. It doesn't just find the lowest price; it calculates the "Total Cost of Acquisition"—factoring in mileage and time—to tell you if driving to a cheaper store actually saves you money.

How we built it

  • Frontend: Built with Vue.js for a reactive, mobile-first mapping experience.

  • Backend: A Python server running on Uvicorn for high-performance asynchronous task handling.

  • Database: MongoDB stores our product catalogs and store locations.

  • Intelligence: Gemini API acts as the core "Reasoning Engine," parsing messy natural language shopping lists into structured data and performing the cost-benefit logic.

Challenges we ran into

The biggest hurdle was data access. High-quality, real-time retail APIs are often locked behind expensive paywalls or strict rate limits. Within the 6-hour sprint, we pivoted to architecting a robust Mock Data Engine in MongoDB that simulates dynamic price fluctuations, allowing us to demonstrate the AI’s decision-making logic without a $500/month API subscription.

Accomplishments that we're proud of

We successfully integrated a "Reasoning Agent." Instead of a basic search filter, our implementation of the Gemini API can understand context—like realizing that "coke" and "soda" are the same intent—and provide a verbal justification for why a specific route is the most economical.

What we learned

We learned that in a hyper-local context, Data Orchestration is more important than Data Ownership. You don't need to own the stores to provide value; you just need to be the smartest layer of logic between the data and the resident.

Built With

Share this project:

Updates