FireSight NYC

In 2022, a fire at the Twin Parks apartments in the Bronx killed 17 people. The doors that should have self-closed had been cited as violations years earlier — but those signals lived in separate city databases that never talked to each other. FireSight asks a blunt question: if the data already existed, could the right building have been inspected first?

• Joined four siloed NYC Open Data sources (via the Socrata API) into a single ontology in Palantir Foundry.
• Built a transparent 0–100 risk score weighting self-closing-door violations, complaint history, and building age — every input visible and defensible, not a black box.
• Designed a three-view operator workflow in Foundry Workshop — an inspection command map, a building detail view, and a historical case study — with AI-generated rationales and one-click dispatch (Python pipeline; AIP Logic for the reasoning layer).

• Using only data available before the fire, the model ranked Twin Parks #1,003 of 89,496 Bronx parcels — the top 1.1%.
• A ranked, explainable inspection queue that turns scattered violations into a defensible order of action.