I’m creating a web-based travel assistant that learns about each visitor through a short, adaptive questionnaire and then recommends their ideal destination. The question flow will probe three areas—preferences and interests, previous travel experiences, and budget plus trip duration—while the backend AI weighs beach escapes, cultural landmarks, and adventure spots equally. Its core objective is to strike a balance between well-known favourites and lesser-known gems, so suggestions always feel both exciting and personalised. Here’s how I picture the build: the front end presents dynamic questions, the answers feed a recommendation engine (open-source LLM or a custom model is fine) that ranks destinations from a curated database, and the results page showcases the top match with supporting information and imagery. An admin area should let me add or edit questions and destinations without touching code, and the entire stack needs clean, well-commented code plus deployment instructions so I can host it on a standard cloud service. Deliverables • Responsive front end with adaptive question flow • Destination database seeded with a diverse mix of popular and offbeat locations • AI recommendation logic that factors in all captured traveller data • Simple admin dashboard for content updates • Setup guide and brief technical documentation If you have experience blending NLP, recommender systems, and travel data, this should feel right in your wheelhouse. Let’s create a tool that sends travellers somewhere they never knew they always wanted to go.