Inspiration
A close relative of mine was buying a three-story building in Bengaluru. They thought it was a great deal, but it turned out to be a mess, cracks in the walls, water pooling everywhere, you name it. That got me thinking why didn't any of the brokers or real estate advisors tell them about these issues and that's when I realised even they didn't know about these issues as there was no software/tool to analyze it.
What it does
BangMaps is an AI agent that analyzes property risks using government datasets and interactive maps, giving insurance providers real-time risk assessments, policy recommendations, and easy integration for underwriting, valuation, and claims processing.
Right now, it’s specially trained on data for Bengaluru houses, looking at things like flood-prone spots, sewage paths, groundwater levels, air quality, slum areas, and more. It analyzes all that to make a risk analysis report for a house in just about no time.
How we built it
We built BangMaps using Next.js for the frontend, Tailwind CSS for styling, mistral, openai apis for function calling, perplexity for web search. ElevenLabs api for text to speech. Backend is built using flask. Leaflet for map visualization, Opencity for datasets related to Bengaluru
Challenges we ran into
We did face some challenges like, dealing with variety of data and processing them altogether.
Accomplishments that we're proud of
The agent works seamlessly.
What we learned
You can just do things
What's next for BangMaps
Ws only
Built With
- elevenlabs
- leaflet.js
- mistral
- openai
- vercel
Log in or sign up for Devpost to join the conversation.