0/60 Wasn't the Model: The Empty Haystack Behind My Two Worst Corpora

Automated excerpt

Fix 1 (#726): the adapters pool was empty. The result, both models, no model change: Corpus v0.3.0 v0.3.1 adapters (n=60) 0 pass (100% matcher_gap) 60/60 pass precision 1.000, recall 0.168 (gpt) / 0.175 (llama) 0 → 100% in one cycle. The reference bucket was generic. 18 generic CI references for a corpus of real, repo-specific CI failures. Authored 48 same-domain CI sibling references (generate_ci_replay_refs.py); the pool grew 540 → 588. A corpus that is 100% matcher_gap should trigger "inventory the reference pool," not "upgrade the model." Authoring same-domain references.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.