Inspiration
The 2026 World Cup is the biggest sporting event in history — 48 teams, 12 venues across 3 countries, millions of fans traveling internationally. Planning a trip to see your team play is overwhelming. I wanted to build something that makes it simple.
What it does
MatchDay AI is a World Cup 2026 trip planning agent. Fans ask natural language questions like "When does Nigeria play?" or "Plan my trip to Dallas for the Argentina match" and the agent instantly pulls real match schedules, venues, ticket prices, and cities from MongoDB to generate a personalized trip plan.
How I built it
- Gemini handles the natural language reasoning and response generation
- MongoDB Atlas stores all match schedules, venues, and team data — the agent queries it in real time
- Python + Gradio powers the web interface
- Render hosts the live application
Challenges
- Connecting MongoDB Atlas from different network environments required careful configuration
- Getting Gemini to ground its responses strictly in the MongoDB data (not hallucinate match details) required careful prompt engineering
What I learned
How to build a real data-grounded AI agent that uses a live database as its source of truth — not just a chatbot making things up.
What's next
Adding hotel booking integrations, flight search, and fan community features.
Log in or sign up for Devpost to join the conversation.