Hyperlocal: Architecture of a System to Connect users with Blue Collar Workers

Urban blue-collar service markets remain highly fragmented, resulting in inefficient worker discovery, inconsistent service quality, and limited trust between customers and service providers. This paper presents Kaargar, a geo-aware hyperlocal service marketplace designed to enable real-time matching between customers and skilled workers such as plumbers, electricians, and technicians. The system integrates geospatial intelligence, transactional integrity, and real-time communication within a modular, scalable architecture. The platform leverages Post-GIS-powered proximity search and availability filtering to enable precise worker discovery within defined geographic radii. A structured job lifecycle engine supports job posting, bidding, assignment, execution, and approval workflows, while an escrow-backed digital wallet ensures secure financial transactions and transparent payment release. Trust and platform governance are strengthened through KYC verification, ratings, complaint resolution, and administrative moderation. To deliver a responsive user experience, the system employs an event-driven architecture using Web-Sockets and Redis-based messaging to provide real-time notifications and job-specific chat. The backend is implemented using FastAPI with asynchronous processing for high concurrency, while Supabase services provide authentication and secure media storage. The proposed architecture demonstrates a scalable and trustworthy digital infrastructure capable of reducing worker discovery time, improving service reliability, and strengthening governance in hyperlocal labor ecosystems.