A fast-fail Python package for LLM-based document reranking with Azure OpenAI, RankGPT, PRP, Setwise Heapsort, TourRank-r, and AcuRank.
-
Updated
Jul 23, 2026 - Python
A fast-fail Python package for LLM-based document reranking with Azure OpenAI, RankGPT, PRP, Setwise Heapsort, TourRank-r, and AcuRank.
Production-grade HR document intelligence system built on n8n, Pinecone, OpenAI, and PostgreSQL. Automatically detects and processes multiple files from a Google Drive folder, then answers natural-language queries against your HR documents with cited, confidence-scored responses — complete with query logging, caching, and error handling.
Sequential recommendation datasets from Amazon Reviews 2023, built for LLM reranker evaluation
TREC ROBUST04 ad-hoc retrieval system achieving MAP 0.4044 via a Seeded LLM Tournament (GPT-4o-mini listwise judge) and a Hybrid BM25+RM3+MonoT5+RRF pipeline. MSc final project, Reichman University.
Defensive publication: two-stage RAG ranker whose LLM rerank stage fails open to a deterministic base.
Semantic candidate search and information retrieval over ~200K profiles: Voyage-3 embeddings, hard/soft relevance filtering, and GPT-4o-mini reranking across 10 role configs. Exhaustive vector-database scans give exact hard-criteria recall, then a blind rubric-calibrated LLM judge reranks finalists. 90.3 avg, 100% hard-criteria pass.
To associate your repository with the llm-reranking topic, visit your repo's landing page and select "manage topics."