Authors

Human author

Brandon Behring — primary author; final responsibility for content, claims, and design.

AI collaboration

This guide follows the AI-collaboration disclosure norms set by ACM 2023, Nature 2023, and COPE 2023. AI tooling is not listed as an author; humans take final responsibility.

For the full workflow disclosure — which skills, which prompts, which review steps — see the hub at guides.brandon-behring.dev/methodology.

Companion + capstone

• companion/ (Python package, mini_eval) — the build-your-own evaluation library you construct chapter-by-chapter (metrics, a mock LLM-as-judge harness, …), bridged to the production tools (eval-toolkit, RAGAS).
• capstone/ — per-guide capstone (the “design an evaluation strategy” mock); authored once the guide’s chapters are complete.

Contributing

PRs welcome for:

• Factual corrections (link to source or counter-example)
• Clarity improvements that don’t change scope or claims
• New examples or worked exercises in the existing chapter shape
• Companion package bug fixes + type annotations

Out of scope for PRs:

• New chapters or major reorganization (open an issue first)
• Style-guide changes (those live in the hub repo)
• AI-tooling reconfiguration (the discipline is intentional)