PizzaTipz

A personal experiment in combining fixture data, Perplexity research, and Claude-generated analysis. These notes describe the pipeline design, not a verified current schedule or profitable selection service.

AI-assisted personal project. Tools and implementation notes are not claims of independent professional specialization or verified live results.

Stack

Engineering highlights

• A proposed discover → research → analyse → build → settle → evaluate loop; feedback is not evidence of improvement by itself.
• Structured LLM orchestration: per-fixture research prompts, constrained JSON outputs, and confidence/odds bounds.
• Settlement and per-leg grading experiments using results APIs, with calibration and drift analysis to be judged against recorded outcomes.