LLM Evaluation

Covers automated output assessment, LLM-as-Judge patterns, and reference-case testing. Specific enough to attract readers searching for evaluation methodology; broad enough to carry forward to any issue covering how AI output gets measured.