A declarative travel planner that starts from a mood rather than a destination list, say wanting to kayak on a quiet lake, and builds an itinerary from it. It reads across popular spots, hour-by-hour crowd data and reviews, then orders stops to dodge the busy hours. Change your mind mid-chat and it recomputes on the spot.
Judges praised the integration of search, routing and RAG, calling the specification and implementation the most substantial in the field. Measuring and improving on accuracy figures marked it out among travel entries as more than an API wrapper.
Travel planners are a crowded category, and standing out means proving you built more than a wrapper around an API. Here search, routing and RAG are pulled into a single flow, and the team measured accuracy and performance and iterated on the numbers, a habit that likely separated it from its neighbours. The conversation matters too: turning a vague mood into a concrete plan, and recomputing when the user changes their mind, ties straight to the payoff of fewer crowds.
See all winners of 第2回 AI Agent Hackathon with Google Cloud →
Summaries are written by this site. Project pages include demo videos, screenshots and the team's own write-up (videos may autoplay).