rank-recon

Audits ranking quality by analyzing metric trends, failure modes, dataset coverage, and reranker performance.

Builds ranking evaluation workflows including NDCG/MRR metrics, human relevance labeling, and offline evaluation harnesses. Useful for search relevance and recommender system testing.

Designs ranking pipelines with reranker selection, score fusion, and cross-encoder patterns. Useful for optimizing search or recommendation ranking with latency trade-offs.