Week 7 · Day 5 — PROJECT: AI Scout Report Generator

Week 7: "Going Pro" | Day 5 of 5 — PROJECT DAY 🏆 Theme: Type any player name → get a full, structured AI scouting report

The Mission

You've spent this week learning to:

• Call the OpenAI API from Python
• Get structured JSON responses
• Build input-to-output pipelines with error handling
• Use Anthropic Claude as an alternative provider
• Compare AI outputs programmatically

Today you're building a production-grade AI Scout Report Generator.

How it works:

1. User types a player name (real or hypothetical)
2. Program asks for their stats (or uses defaults)
3. Calls AI API to generate a complete scouting report
4. Formats and saves the report
5. Optionally runs comparison analysis with a second model

What makes this "production-grade":

• Input validation at every step
• Retry logic on API failures
• Structured JSON parsing with fallbacks
• Multiple output formats (console, markdown, JSON)
• Saved to disk
• Comparison mode with two AI models