NestBuilder

Explore
ViewNest

🏡 NestBuilder

Inspiration

Moving to a new city is overwhelming. You don’t know where to find your favorite coffee shops, gyms that match your workout style, or restaurants that serve the cuisine you love. I built NestBuilder after experiencing this frustration firsthand—starting over in an unfamiliar place, not knowing where anything is, and spending weeks trying to find my “spots.”

What if AI could analyze your existing preferences and instantly recommend places in your new city that match your lifestyle?

What it does:

NestBuilder helps people moving to new cities feel at home faster through AI-powered personalization.

Analyzing Your Preferences Upload your Google Maps Takeout data, and our AI (powered by Google Gemini) analyzes your search history to understand your lifestyle habits and interests.

AI-Powered Onboarding Automatically infers: Home and work locations Transportation preferences (walking, biking, driving, transit) Categories you care about (restaurants, gyms, parks, cafés, etc.)

Personalized Recommendations: For each category, Gemini generates tailored recommendations based on: Your historical habits Distance from your new home Environment descriptors (cozy, upscale, casual, quiet, etc.)

My Nest Save favorite discoveries to a personal collection featuring: Interactive maps Marker tooltips Click-to-zoom functionality Google Maps integration

How we built it

Frontend: React 18 with TypeScript Material UI (MUI) Firebase Authentication & Firestore

Backend: Python Flask REST API Google Gemini AI (gemini-3 AI) for lifestyle analysis Integrations Google Maps JavaScript API Google Places API Google Distance Matrix API

Data Processing Custom parsing of Google Takeout data Identification of multi-modal search patterns and lifestyle habits

Challenges we ran into

API model updates — Migrating from deprecated gemini-1.5-pro to gemini-2.0-flash

Google Places legacy issues — Enabling legacy APIs for photo and place detail support

State management — Fixing stale closure issues in React’s “Load More” functionality

Home address extraction — Prompting AI to recognize explicit search patterns to infer home/work locations

Accomplishments that we’re proud of:

Seamless AI integration: Structured lifestyle profiles from free-form search history

Real personalization : Recommendations based on actual habits, not generic lists

Interactive UX: Smooth mapping with hover details, click-to-zoom, and saved locations

Clean architecture: Strong separation between frontend and backend APIs

What we learned

Prompting Gemini AI for reliable structured JSON outputs Secure OAuth token and API key management Building high-performance interactive maps Real-time data synchronization between Flask and Firebase