Inspiration
Our inspiration came from navigating the neighborhoods and cities of LA County and Orange County as university students. The sheer number of local communities is outstanding yet overwhelming from a travel and real estate perspective. From weekend trips around SoCal to scoping out the viability of apartments near school, ZipScope has you covered.
What it does
ZipScope is an interactive webpage that provides users with intuitive insight into the living desirability of any address or zip code across Los Angeles and Orange Counties. It generates a comprehensive desirability score out of 100, based on six key metrics: Safety, Housing Affordability, Public Education Quality, Transportation, Environment, and Healthcare Accessibility. Each metric is visually represented on the platform, which also includes a feature allowing users to easily compare and assess various locations. By combining data-driven analysis with an engaging interface, ZipScope empowers users to make informed decisions about the best places to live, work, or invest.
How we built it
- We began by initializing our app with a Next.js/React front-end (Tailwindcss styling) and a FastAPI back-end.
- We sourced data for crime, pollution, health accessibility, education opportunities, etc. in the LA/OC area from various APIs and public datasets, including the Melissa ZIP Code dataset. We analyzed the data to determine the best metrics for our program.
- We integrated Meta's Llama 3.3-70B model (run on Groq) to process dynamic geolocation queries, synthesizing multi-dimensional data to generate scores.
Challenges we ran into
- Tailwindcss + Leaflet.js integration difficulties (Leaflet map is extremely sensitive to element + CSS nesting).
- Integration among different devices.
- Front-end accessibilities for users.
Accomplishments that we're proud of
- Successfully integration for LLM's to power our program.
- All encompassing UI to guide a user through a scope process.
- Applying new technologies for a cohesive experience.
What we learned
- Optimizing AI models like Meta’s Llama 3.3-70B to process diverse datasets efficiently.
- Designing intuitive, accessible front-end solutions for a seamless user experience.
- Collaborating effectively to integrate complex technologies into a cohesive project.
What's next for ZipScope
- Expanding to neighborhoods beyond the OC/LA area.
- More comprehensive reports for specific metrics.
- Launching a reverse search method to allow users to select their needs and we will return several matches.
Built With
- fastapi
- groq
- html
- javascript
- leaflet.js
- llama-3.3
- melissa-data
- nextjs
- python
- react
Log in or sign up for Devpost to join the conversation.