MovingTUMorrow 🏡🚀
Inspiration 💡
Finding a home is often more complex than just picking a place—it's about envisioning your future. Whether it's ensuring proximity to schools for your kids, a gym for your hobbies, or planning a mortgage, the process can feel daunting. We wanted to cut through the noise and create a smarter, simpler, and more personalized way to navigate housing decisions.
What it does ⚡
MovingTUMorrow combines a complex interaction graph of AI agents with an intuitive user interface to simplify housing and financial decisions:
Interactive Life Planner
- Helps users simulate different life scenarios and paths
- Aligns housing needs with future plans through human-in-the-loop feedback
- Allows modification and selection of proposed scenarios
Smart Housing Matches
- Provides tailored recommendations based on:
- User preferences and budget
- Proximity to schools, family, sports facilities
- Cultural and lifestyle activities
- Match between house features and life planning choices
- Provides tailored recommendations based on:
Interactive Map Visualizer
- Displays housing options with detailed information markers
- Shows distances to all places of interest
- Updates dynamically based on life planning choices
Finance Advisor 💰
- Connects directly to banks via PSD2
- Automates financial data processing
- Provides personalized mortgage options
How we built it 🛠️
- AI Core: LangChain graphs with Google Vertex AI for interactive decision-making
- Data Integration: Google Maps API, ThinkImmo Business API, ImmoScout24 API
- Finance: PSD2 integration for automated financial processing
- Frontend: Interactive map visualizer using Google Maps API
Challenges we ran into 🤔
- API Reliability: The provided APIs had limitations we needed to work around
- ThinkImmo API works only for new housing data polling
- ImmoScout24 required paid access, leading to mockup development
- Interface Design: Creating simple workflows for complex processes
Accomplishments 🏆
- Built a user-centric platform connecting housing needs with data and financial realities
- Developed a scalable solution that can be easily extended to other cities
- Created a clear, multi-step process that makes house hunting engaging
What we learned 📚
- Personalization is key - housing isn't one-size-fits-all
- Integrating multiple data sources creates powerful synergies
- Frontend development skills (from a team of ML engineers!) 🤓
What's next 🔮
- API Integration: Implementing full API functionality
- City Expansion: Adding more German cities and regions
- Financial Tools: Adding tax benefits, renovation costs, and savings calculators
- Gamified Experience: Adding progress badges and milestones
This project is our contribution to improving the housing crisis one small step at a time.
Built With
- fastapi
- google-cloud
- google-maps
- langchain
- python
- react
- typescript
- vertex-ai
Log in or sign up for Devpost to join the conversation.