Dev Log
Education
Building Zestly
Find the perfect school for your child.
Architecture
Zestly uses a personality-based matching system. Parents answer questions about their child's grade level, personality, and interests. Claude AI analyzes the profile against official NCES (National Center for Education Statistics) data — enrollment, demographics, programs, and activities — to generate personalized school recommendations with clear reasoning. A search-by-location fallback lets parents browse directly. Parent guides provide school visit checklists, enrollment timelines, and transition tips.
Tech Stack
Find Schools
zestly.school/
Build Timeline
NCES data pipeline + school profiles
Built the data pipeline ingesting official school data from the National Center for Education Statistics. Each school profile includes enrollment numbers, demographics, grade levels, and program offerings. Profiles are statically generated with ISR for fast loading across all K-12 schools.
AI-powered matching with Claude
Integrated Claude AI for personalized school recommendations. The matching flow asks parents about grade level (Elementary, Middle, High School), then deeper questions about the child's personality and interests. Claude analyzes the profile against school data and provides recommendations with clear reasoning for each match.
Activities matching + parent guides
Added activities and programs matching — sports, arts, STEM clubs, and more — connecting each child's interests with schools that offer them. Built the parent guide section with school visit checklists, enrollment timelines, and tips for navigating the process. Added search-by-location as a browse fallback.