Recipe · Search & RAG
Legal passage ranking
Use Jev after an initial document search.
Open docs.typesafe.ai ↗How builders describe it
One question per query-candidate pair raises top-1 accuracy from 5% to 18% and top-10 accuracy from 38% to 62% on 40 CLERC legal queries.
The decision Jev makes
Rank candidates against a query, or answer whether a passage is relevant.
Where it fits
Search pipelines retrieve too much. These projects use Jev as a reranker or relevance check after the first search, so only useful results reach the user or the model. All 108 search & rag projects →
Related projects
Search & RAGjev-keyboardJev-powered candidate reranking for Rime on macOS, with Windows support planned.GitHub · ★ 4 · Ideny42
Search & RAGSmart FindCtrl+F, but for questions.GitHub · ★ 1 · eduardoalba00
Search & RAGRepoFinderFind your way around a large open-source repo: architecture map plus "where is X implemented?" answers with line citations; optional TypeSafe Jev…GitHub · SelmanBoraySearch & RAGagent-doc-searchPoint coding agents at the right docs in large codebases: local BM25 and a TypeSafe Jev Choice search propose documents, then a Jev yes/no check…GitHub · beqauxSearch & RAGFilmHunchSelf-hostable movie finder: describe a film any way you like and TypeSafe Jev scores film cards and full profiles against your exact sentence.GitHub · eduardoalba00
Search & RAGVector Graph RAGGraph RAG framework over Milvus vector search with no graph database required: Jev Noul judgments select graph relations and rerank full source…GitHub · ★ 258