Secure Ride-Hailing App for Women

Abstract

The safety of women using ride-hailing services in Pakistan remains a major concern, with frequent reports of harassment, route deviations, and low trust due to mostly male drivers and weak security protocols on platforms like Careem, InDriver, and Yango. Existing safety features are reactive and fail to prevent incidents or address women’s specific vulnerabilities. This project fills that gap by developing and evaluating “Golabo,” a proof-of-concept Secure Ride-Hailing App exclusively for women. Golabo replicates standard ride-hailing features and layers in proactive security measures: mandatory per-ride biometric face verification with SFace and MiniFASNetV2 liveness detection to prevent impersonation; continuous, real-time monitoring via a .NET Insight Engine of GPS data, device sensors, and speed to flag route deviations, prolonged stops, or dangerous driving; and a dual real/fake PIN system for discreet distress signaling to a 24/7 monitoring team. Registration is female-only via a Gender Detector, and document uploads undergo antivirus scanning with ClamAV. Implemented as a containerized, multi-tier microservices architecture, Golabo’s backend includes a .NET 9 API, a Python/FastAPI AI image-processing service, the Insight Engine, PostgreSQL (EF Core Code-First), and ClamAV. The front end features a Flutter mobile app for users and Next.js portals for monitoring, all communicating over HTTPS and WebSockets, orchestrated by Docker Compose behind NGINX. Security is enforced throughout with JWT, role-based authorization, custom middleware, input validation, salted hashing, and AES encryption. This proof-of-concept demonstrates the feasibility and value of integrating multi-layered, proactive security into ride-hailing platforms, offering a technical blueprint for enhanced, verifiable safety measures tailored to women and a model for raising industry standards.