Application Development (Web / mobile / Desktop)

Abstract

This thesis presents Veltra Stock, a cloud-based inventory management platform designed for small grocery retailers. It addresses the common challenge of limited access to affordable and user-friendly systems that also provide intelligent decision support. The platform stores per-user inventory data in Google Cloud Firestore and integrates a backend-managed Google Gemini assistant for natural-language insights. To improve trust and relevance, AI responses are grounded in structured inventory snapshots and protected by scoped data access controls. Overall, the study demonstrates that combining secure cloud architecture with explainable AI can improve stock visibility, usability, and day-to-day decision-making in small retail operations.