Article’s

In the contemporary landscape of digital governance, the equitable delivery of state-sponsored welfare schemes to intended beneficiaries remains a persistent and structurally complex challenge, particularly within demographically diverse and socio-economically stratified nations. Despite the conceptualization and robust fiscal allocation of numerous governmental welfare programs, the marginalized populations they are specifically designed to serve are frequently excluded from their benefits due to compounding systemic barriers: pervasive information asymmetry, convoluted eligibility documentation requirements, the absence of proactive eligibility inference tools, and a profound lack of spatial intelligence available to field intervention agencies. This research presents Sabal, a unified, cloud-native, dual-platform technological ecosystem engineered to systematically dismantle these intersecting barriers through the application of modern artificial intelligence, asynchronous web architecture, and geospatial analytics. The Sabal ecosystem comprises two architecturally distinct yet data-synchronized platforms. The first, Sabal Setu, is a citizen-centric portal that employs a rule-based demographic eligibility engine to proactively match individual citizens against a structured repository of over fifty-seven active governmental schemes. It further incorporates a multimodal AI document vault powered by the Google Gemini 2.0 Flash API to automate the extraction of identity and demographic data from uploaded civic documents, eliminating the manual transcription bottleneck that causes a disproportionate rate of application rejection among low-literacy demographics. The second platform, Sabal AI, functions as an enterprise-grade intelligence dashboard designed for Non-Governmental Organizations and administrative coordinators. It leverages interactive geospatial mapping via a react-leaflet rendering engine to visualize civic service gaps across geographic zones, and computes a proprietary Social Return on Investment metric to empirically optimize on-ground resource deployment decisions. Field directives are further enriched by a generative AI layer that synthesizes zone-specific demographic statistics into actionable natural-language Field Intelligence Briefs. The complete system is built upon a highly scalable monorepo framework utilizing React 18 and TypeScript on the frontend, Node.js with Express.js on the backend, and PostgreSQL managed through the Prisma ORM for type-safe, ACID-compliant relational data persistence. Security across the platform is enforced via JWT stateless authorization and bcryptjs cryptographic password hashing. Experimental evaluation using synthetic demographic datasets demonstrates the ecosystem’s capacity to accurately surface eligible scheme recommendations, autonomously extract structured identity data from heterogeneous civic documents, and generate empirically ranked geographic intervention priority lists for NGO deployment. This paper presents the complete architecture, computational methodology, and functional outcomes of the Sabal ecosystem, advancing a scalable and empirically grounded paradigm for next-generation inclusive digital governance.