A team of COMP Year 4 students has proudly received the Best Innovation Award in the Tertiary Stream at the Open Data Hackathon 2025 (Smart Economy), organised by the Hong Kong Technology Advancement Group (HKtag), following the grand finals held in January 2026. Team Members: LI Kwan To (Leader), LI Chun Fung and HO Cheuk Ting Supervisor: Prof. NG Hiu Fung Peter, COMP Assistant Professor Project: HKAssist | A City-Level AI Coordination Layer for Public Services Designed to address the fragmentation of public services in Hong Kong, HKAssist acts as a unified integration layer built on top of existing government infrastructure. By combining Large Language Models (LLM) and Retrieval-Augmented Generation (RAG) with workflow orchestration, the platform provides a conversational interface that helps citizens, SMEs, and visitors navigate cross-departmental procedures. Instead of just offering information, HKAssist guides users seamlessly from initial enquiry to task completion—handling requirement identification, document preparation, and progress tracking. This transforms fragmented data into actionable processes, bridging the gap between "knowing" and "doing" while maintaining transparency and human-in-the-loop control. Perfectly aligning with the hackathon's "Smart Economy" theme, HKAssist also features a 3D Gaussian Splatting (3DGS) experiential module dedicated to tourism. This web-based feature enables immersive, 360-degree virtual exploration of Hong Kong locations without specialised hardware. It encourages visitors to generate and share interactive XR content, transforming individual experiences into long-term cultural and data assets for the city. By balancing operational efficiency in administrative services with engaging experiential features in tourism, HKAssist showcases the future of smart city innovation.
