AI Assistant for Resource Directories

Description
Developed a prototype of an ai assistant (Hope) for resource directories. A multilingual AI assistant designed to simplify access to community resources like food, housing, and healthcare. The solution is targeted at community-based organizations (CBOs) that currently rely on static, outdated directories that frustrate families and erode trust.

Challenge / Problem
Families often abandon resource directories after failed searches or confusing terminology. Information is frequently outdated or incomplete, leading to poor matches and diminished trust. Meanwhile, CBOs lack visibility into what families are asking for, where gaps exist, and how resources are performing.

What We Did
Through user discovery with families and nonprofit leaders, we identified key barriers and prototyped Hope, an AI assistant that:
• Understands natural language requests (“I need food tonight” vs. “emergency pantry”).
• Offers 24/7 multilingual support.
• Uses Retrieval Augmented Generation (RAG) to ground answers in vetted resource data.
• Integrates with Model Context Protocol (MCP) for real-time updates and verification.


Impact / Insights (Pre-Launch)
• Conversational AI shows strong potential to reduce abandonment by guiding families directly to resources.
• Early prototypes indicate meaningful time savings for both families and staff.
• Nonprofits see opportunities to unlock insights into trending needs, service gaps, and funding priorities.