Building a RAG-Powered Hackathon Winning App
SMRTR summary
Hangout AI, a personalized travel itinerary generator, placed fifth at the TiDB Future App Hackathon 2024. It uses Retrieval-Augmented Generation (RAG) and vector search to create custom travel plans based on user input, location, date, and weather. The system architecture includes data scraping, vector retrieval using TiDB, and itinerary generation with the Groq llama3-70b-8192 LLM. While promising, Hangout AI requires improvements in accuracy, error handling, security, and scalability before becoming production-ready.
SMRTR provides this summary for quick context. The original article belongs to GitConnected.
Read the original article