The original title is: Building "PitchPassion AI": Transforming Football Passion into Data with Google AI

Short summary

A weekend hackathon submission called PitchPassion AI that uses Google Gemini API to extract player performance data from unstructured football match narratives and visualize it as charts. The app is built with React and Flask, with middleware to clean Gemini's JSON output and retry logic for API stability issues. The author highlights prompt engineering and backend sanitization as key problem-solving strategies.

• •PitchPassion AI converts football match narratives into player performance visualizations using Google Gemini
• •Built with React, Flask, and Gemini API for NLP-based structured data extraction
• •Key challenges solved: API 503 errors via retry logic and inconsistent JSON formatting via prompt engineering and sanitization

Generated with AI, which can make mistakes.