MetroCore Administration

MetroCore Administration is a São Paulo-based corporate civic-AI firm operating from a building in the Vila Olímpia district that the firm purchased outright six years ago, a decision the founders describe as “institutional commitment to São Paulo.” Three hundred and fifty staff, a client roster of Brazilian and Latin American megacity administrations, and a corporate culture that the firm’s HR team has internally described as “South American institutional with North American operational discipline,” which translates, in practice, to relationships that take months to build and decisions that get made quickly once they are.

Their work is corporate-tier governance AI for megacity-scale civic administration — resource allocation across district populations in the millions, infrastructure-stress modeling, public-services demand prediction for the kind of metropolitan environments where the wrong allocation decision becomes visible on the evening news within hours. Per-engagement Cu invoicing settles in the mid seven-figure range; full city-scale deployments clear higher. The firm’s contracts are negotiated patiently. The implementation runs are managed precisely.

The São Paulo office contains, on its executive floor, a wall-sized real-time visualization of the city’s traffic, transit, and emergency-response patterns — a deployment of an early version of the firm’s own product, run continuously since the firm’s founding. The display has, the firm’s CEO has noted in interviews, “shaped every product decision since.”

MetroCore Administration’s senior research team has, for the past several years, maintained an internal project they refer to only as “the comparison study.” The project examines patterns in citizen-data deployments across the firm’s client cities and produces findings about, as the senior researcher has carefully phrased it in a closed-doors industry presentation, “the things urban populations do when they realize they are being measured.” The findings are not published. The clients are not informed. The senior researcher has, on several occasions, presented selected findings to small groups of academic peers under non-disclosure. The peers have, in turn, declined to publish.