layout: project title: Yinzone description: Operating System powering the burgh's affordable housing build chain for developers and zoning regulators. image: /assets/images/projects/yinzone.webp card_image: /assets/images/projects/cards/yinzone-card.webp start_date: 2026-09-26 end_date: 2026-09-27 activity_status: production links:
- title: yin.zone url: https://yin.zone icon: fa-solid fa-up-right-from-square
- title: Watch the demo url: https://www.youtube.com/watch?v=ECZnv7R-q_8 icon: fa-solid fa-play
- title: src url: https://github.com/matmanna/yinzone icon: fa-brands fa-github
- title: Pittsburgh AI Horizons url: https://ai-horizons-2026-ai-for-housing-hackathon.brandon831577.chatgpt.site/ icon: fas fa-trophy tags:
- GIS
- Zoning
- React
- Gemini
- Hackathon skills:
- name: Cartography icon: fa-solid fa-compass-drafting
- name: Data Analysis icon: fa-solid fa-diagram-project
- name: Full-stack Web icon: fa-solid fa-layer-group
Yinzone is a parcel-level decision-support map for the City of Pittsburgh, built at the AI Horizons 2026 hackathon (Track 3 — Housing Typology, Equity & Climate Matchmaker).
Click any of ~142,000 parcels and see its overall 0–100 score, what the zoning code allows there for 16 housing types and by which approval pathway, whether the site physically fits, a rough "does it pencil?" cost screen, and every source behind each number. Deterministic code decides every score; the AI (Gemini chat + a typed site-fit model) only explains or rates from pre-computed facts, never fabricates a number.