Inspiration
The global housing crisis affects over 1.6 billion people, with millions living in informal settlements exposed to floods, poor air quality, and unsafe waste systems. Families spend more than 22 weeks searching for safe housing, often with only a 23% success rate. We were inspired to leverage AI and climate intelligence to cut through this inefficiency, so no family is forced to choose between affordability and safety.
What it does
HomeHarbor is an AI-powered platform that transforms housing access. It analyzes environmental risks such as flood zones, air quality, waste systems alongside affordability and availability to provide climate-resilient housing matches in just 6 weeks instead of 22. With multilingual AI support and SMS notifications, HomeHarbor works globally, connecting families from urban U.S. neighborhoods to informal settlements in Nairobi to safe homes faster, smarter, and more sustainably.
How we built it
IBM watsonx for geospatial AI analysis of flood, air, and climate data.
IBM Watson NLP for multilingual chat support across 15+ languages.
IBM Watson Discovery to process global housing databases.
IBM Cloud Functions for scalable SMS notifications in areas with limited connectivity.
IBM Environmental Intelligence Suite to evaluate environmental safety metrics.
OCR & workflow automation for user document uploads, affordability checks, and offline support.
Challenges we ran into
Integrating diverse environmental datasets from satellite imagery, air quality indices, and flood maps into a single, real-time analysis pipeline.
Ensuring the platform remained accessible in low-connectivity regions while maintaining global scalability.
Designing AI governance and ethical safeguards to prevent bias in housing recommendations.
Balancing technical depth with usability, ensuring the tool is both powerful and simple for families under stress to use.
Accomplishments that we're proud of
Reduced housing search time from 22+ weeks to just 6 weeks.
Increased housing placement success rates from 23% to 87%.
Built a multilingual system supporting 15+ languages for inclusivity.
Demonstrated global impact potential, serving both U.S. families and informal settlement communities abroad.
Developed a scalable workflow that integrates AI climate analysis, affordability, and SMS outreach.
What we learned
How to integrate AI tools like Watson, watsonx, NLP, Discovery into a cohesive solution for social good.
The importance of combining climate resilience with housing equity safe housing must also be environmentally sustainable.
That accessibility through offline access, and multilingual support is as crucial as technical performance for real-world adoption.
Collaboration across technical, social, and policy domains is key to solving global challenges like housing inequality.
What's next for Home habour
Scaling partnerships with NGOs, housing authorities, and informal settlement upgrade programs.
Expanding environmental data coverage on wildfire risk, heatwaves, and drought zones).
Adding predictive climate modeling to future-proof housing recommendations.
Deploying mobile-first versions for communities where smartphones are the primary internet device.
Establishing pilots in the U.S., Kenya, and Southeast Asia to refine workflows across diverse contexts.
Log in or sign up for Devpost to join the conversation.