THE COMMUNITY RESOURCE
DIRECTORY
Small decisions.
Small decisions.
Endless possibilities.
Find what people build with Jev. Explore tools, libraries, experiments and ideas from across the TypeSafe ecosystem.
11,066unique resources
29clear categories
220focused subcategories
Yourssave a personal reading list
YOUR NEXT BUILD STARTS HERE
Download catalog
Explore the collection
Browse groups
Search, retrieval & RAG193 of 193 matches
RAG & document retrieval3 of 3 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| jev-raggithub.com | — | Data & Search: Local knowledge search with three modes: BM25 + Jev by default, agentic multi-query lexical search + Jev without vectors, and optional BM25 + embeddings/RRF + Jev, with grounded streaming answers. | Search, retrieval & RAGRAG & document retrieval | |
| jselectgithub.com | 3 | Python context selector that asks Jev whether each passage in your files or records is useful evidence for a task, then picks diverse verbatim passages with source links within a token budget. | Search, retrieval & RAGRAG & document retrieval | |
| Qualitative Querygithub.com | 1 | Obsidian plugin that answers saved questions with original passages from your notes, then saves them into a note linked back to the sources, without generating an answer. | Search, retrieval & RAGRAG & document retrieval |
Ranking & reranking50 of 50 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| Boardy match rerankingx.com | — | Boardy, an AI superconnector, reports that Jev has already reranked over 1 million people-matches in production. | Search, retrieval & RAGRanking & reranking | |
| Boardy match reranking — appboardy.ai | — | Boardy, an AI superconnector, reports that Jev has already reranked over 1 million people-matches in production. | Search, retrieval & RAGRanking & reranking | |
| chia1104.dev Jev rerankergithub.com | — | Search reranker in the chia1104.dev personal site and CMS monorepo that makes one Jev call per query: a Choice over candidate posts and a Noul on whether any candidate actually answers it, working across languages. | Search, retrieval & RAGRanking & reranking | |
| chia1104.dev Jev reranker — appchia1104.dev | — | Search reranker in the chia1104.dev personal site and CMS monorepo that makes one Jev call per query: a Choice over candidate posts and a Noul on whether any candidate actually answers it, working across languages. | Search, retrieval & RAGRanking & reranking | |
| chia1104.dev Jev reranker — repogithub.com | — | Search reranker in the chia1104.dev personal site and CMS monorepo that makes one Jev call per query: a Choice over candidate posts and a Noul on whether any candidate actually answers it, working across languages. | Search, retrieval & RAGRanking & reranking | |
| decision-model-playgroundgithub.com | 1 | Local browser playground where Laya or Jev picks the best-matching vervecode.dev article for a question, showing every option's raw probability, a no-match option and load and inference timings. | Search, retrieval & RAGRanking & reranking | |
| docbank Jev rerankinggithub.com | — | Local-first document vault for people and agents that adds a TypeSafe Jev reranking provider for search over document names and extracted text. | Search, retrieval & RAGRanking & reranking | |
| docbank Jev reranking — appdocbank.ai | — | Local-first document vault for people and agents that adds a TypeSafe Jev reranking provider for search over document names and extracted text. | Search, retrieval & RAGRanking & reranking | |
| endpointgithub.com | 8 | WiktorB2004/llama-index-jev — LlamaIndex reranker + router powered by TypeSafe Jev — typed scores/choices, cheaper than LLM-as-judge. · endpoint · Python | Search, retrieval & RAGRanking & reranking | |
| endpointgithub.com | 10 | hotchpotch/jev-reranker — Jev-powered relevance filtering and reranking for RAG in Python. · endpoint · Python | Search, retrieval & RAGRanking & reranking | |
| hev rerankergithub.com | 14 | Recipe and 90-line Python wrapper that uses Jev as a calibrated reranker for up to 30 documents per call, with nDCG@10 results against hosted and open rerankers on BEIR shortlists. | Search, retrieval & RAGRanking & reranking | |
| Hindsight Jev rerankergithub.com | ≈24,900 | Reranker provider in the Hindsight agent-memory system that asks Jev one question with every recall candidate as an option, so the answer is the ranking, in a single request. | Search, retrieval & RAGRanking & reranking | |
| Hindsight Jev reranker — docsgithub.com | ≈24,900 | Reranker provider in the Hindsight agent-memory system that asks Jev one question with every recall candidate as an option, so the answer is the ranking, in a single request. | Search, retrieval & RAGRanking & reranking | |
| Jev Rerankergithub.com | 4 | Rust CLI, installed from npm, that reads a JSON array of search results from stdin and uses Jev to rerank them, drop candidates with no usable evidence, or extract query-specific passages for your LLM. | Search, retrieval & RAGRanking & reranking | |
| Jev Reranker — articlenorsica.jp | — | Rust CLI, installed from npm, that reads a JSON array of search results from stdin and uses Jev to rerank them, drop candidates with no usable evidence, or extract query-specific passages for your LLM. | Search, retrieval & RAGRanking & reranking | |
| jev-rerankergithub.com | 10 | Python library that uses Jev to score retrieved documents for usefulness as answer evidence, rerank them and drop those under a threshold before generation, handling concurrency, long candidate lists and retries. | Search, retrieval & RAGRanking & reranking | |
| jev-reranker — demox.com | — | Python library that uses Jev to score retrieved documents for usefulness as answer evidence, rerank them and drop those under a threshold before generation, handling concurrency, long candidate lists and retries. | Search, retrieval & RAGRanking & reranking | |
| jev-reranker — write-uphuggingface.co | — | Python library that uses Jev to score retrieved documents for usefulness as answer evidence, rerank them and drop those under a threshold before generation, handling concurrency, long candidate lists and retries. | Search, retrieval & RAGRanking & reranking | |
| jev-search by kylemclarengithub.com | 8 | A shadcn/ui registry block: keyword hits on the first keystroke, re-ranked by Jev a moment later, and keyword order stands if the call fails. | Search, retrieval & RAGRanking & reranking | |
| jev-siftgithub.com | 46 | Agent plugin and MCP tool that sends batches of files, public web pages, text, or tool descriptions to Jev with a query and returns relevance probabilities, so the agent opens only items worth reading. | Search, retrieval & RAGRanking & reranking | |
| jev-sift — demox.com | — | Agent plugin and MCP tool that sends batches of files, public web pages, text, or tool descriptions to Jev with a query and returns relevance probabilities, so the agent opens only items worth reading. | Search, retrieval & RAGRanking & reranking | |
| jevsearchgithub.com | 8 | Drop-in shadcn/ui block for site search that shows keyword hits on the first keystroke, then re-ranks them by the visitor's intent with Jev a moment later, keeping keyword order if the call fails. | Search, retrieval & RAGRanking & reranking | |
| jevsearch — appjevsearch.fly.dev | — | Drop-in shadcn/ui block for site search that shows keyword hits on the first keystroke, then re-ranks them by the visitor's intent with Jev a moment later, keeping keyword order if the call fails. | Search, retrieval & RAGRanking & reranking | |
| keep.md search and taggingx.com | — | Test of Jev on Cloudflare Workers inside keep.md: search reranking 7x faster than the current hybrid, and content tagging 50x faster than GLM 4.7 Flash with no failures. | Search, retrieval & RAGRanking & reranking | |
| keep.md search and tagging — appkeep.md | — | Test of Jev on Cloudflare Workers inside keep.md: search reranking 7x faster than the current hybrid, and content tagging 50x faster than GLM 4.7 Flash with no failures. | Search, retrieval & RAGRanking & reranking | |
| Kody Jev search rerankgithub.com | — | Optional second stage for Kody's MCP search that widens hybrid lexical and vector recall on ambiguous queries, then reranks and filters candidates with Jev via Workers AI. | Search, retrieval & RAGRanking & reranking | |
| Kody Jev search rerank — appkody.codes | — | Optional second stage for Kody's MCP search that widens hybrid lexical and vector recall on ambiguous queries, then reranks and filters candidates with Jev via Workers AI. | Search, retrieval & RAGRanking & reranking | |
| Kody Jev search rerank — demox.com | — | Optional second stage for Kody's MCP search that widens hybrid lexical and vector recall on ambiguous queries, then reranks and filters candidates with Jev via Workers AI. | Search, retrieval & RAGRanking & reranking | |
| Kody Jev search rerank — repogithub.com | — | Optional second stage for Kody's MCP search that widens hybrid lexical and vector recall on ambiguous queries, then reranks and filters candidates with Jev via Workers AI. | Search, retrieval & RAGRanking & reranking | |
| LanceDB rerankergithub.com | — | Asks one Noul per query-document pair and stores an absolute relevance probability you can threshold. | Search, retrieval & RAGRanking & reranking | |
| LanceDB reranker — docslancedb.com | — | Asks one Noul per query-document pair and stores an absolute relevance probability you can threshold. | Search, retrieval & RAGRanking & reranking | |
| Lattice Jev rerankergithub.com | 15 | Optional reranking stage in Lattice, a local-first retrieval engine for markdown knowledge bases, that sends the fused top hits to Jev in one request and scores each by the probability it answers the query. | Search, retrieval & RAGRanking & reranking | |
| Lattice Jev reranker — appzabaca.com | — | Optional reranking stage in Lattice, a local-first retrieval engine for markdown knowledge bases, that sends the fused top hits to Jev in one request and scores each by the probability it answers the query. | Search, retrieval & RAGRanking & reranking | |
| OpenViking reranker — appopenviking.ai | — | Rerank provider for a context database that scores every document with a Noul in one request. | Search, retrieval & RAGRanking & reranking | |
| sdkgithub.com | 14 | hev/reranker — Use Jev (TypeSafe's System One model) as a calibrated reranker: one call, up to 30 documents, a probability per document. Apache-2.0. · sdk · Python | Search, retrieval & RAGRanking & reranking | |
| SiftRankgithub.com | 202 | CLI and agent skill for finding the most relevant items in large datasets by iterative ranking, which added a --provider jev option alongside Chat Completions APIs. | Search, retrieval & RAGRanking & reranking | |
| SiftRank — demox.com | — | CLI and agent skill for finding the most relevant items in large datasets by iterative ranking, which added a --provider jev option alongside Chat Completions APIs. | Search, retrieval & RAGRanking & reranking | |
| TrainLCD Jev supportgithub.com | — | Jev integration in the Cloudflare Worker behind the TrainLCD transit app: a per-candidate Noul reranks the stations its AI chat suggests, and a separate module triages user feedback by spam, category and priority. | Search, retrieval & RAGRanking & reranking | |
| TrainLCD Jev support — prgithub.com | — | Jev integration in the Cloudflare Worker behind the TrainLCD transit app: a per-candidate Noul reranks the stations its AI chat suggests, and a separate module triages user feedback by spam, category and priority. | Search, retrieval & RAGRanking & reranking | |
| TrainLCD Jev support — repogithub.com | — | Jev integration in the Cloudflare Worker behind the TrainLCD transit app: a per-candidate Noul reranks the stations its AI chat suggests, and a separate module triages user feedback by spam, category and priority. | Search, retrieval & RAGRanking & reranking | |
| Turbo Rerankgithub.com | — | Search workspace demo that reranks 50 candidates in one ~170 ms Jev request; on 40 queries the author labeled over a generated 598-passage corpus, top-1 rose from 50% to 100% (98% in a browser rerun). | Search, retrieval & RAGRanking & reranking | |
| Turbo Rerank — repogithub.com | — | Search workspace demo that reranks 50 candidates in one ~170 ms Jev request; on 40 queries the author labeled over a generated 598-passage corpus, top-1 rose from 50% to 100% (98% in a browser rerun). | Search, retrieval & RAGRanking & reranking | |
| typesafe-tools Codex plugingithub.com | — | Codex plugin with an MCP server and skill that use bounded Jev evaluations to rank public research passages and check whether sources support a claim, behind spending caps and pilot gates. | Search, retrieval & RAGRanking & reranking | |
| typesafe-tools Codex plugin — repogithub.com | — | Codex plugin with an MCP server and skill that use bounded Jev evaluations to rank public research passages and check whether sources support a claim, behind spending caps and pilot gates. | Search, retrieval & RAGRanking & reranking | |
| Upweightx.com | — | Hacker News front page you re-rank with six sliders (technical depth, drama, practical utility, AI slop, novelty, career relevance), with each story's Jev scores shown; Firecrawl does the reading. | Search, retrieval & RAGRanking & reranking | |
| Upweight — appupweight.vercel.app | — | Hacker News front page you re-rank with six sliders (technical depth, drama, practical utility, AI slop, novelty, career relevance), with each story's Jev scores shown; Firecrawl does the reading. | Search, retrieval & RAGRanking & reranking | |
| WiktorB2004/llama-index-jevgithub.com | 8 | LlamaIndex reranker + router powered by TypeSafe Jev - typed scores/choices, cheaper than LLM-as-judge. | Search, retrieval & RAGRanking & reranking | |
| WiktorB2004/llama-index-jev websitewiktorb2004.github.io | — | LlamaIndex reranker + router powered by TypeSafe Jev - typed scores/choices, cheaper than LLM-as-judge. | Search, retrieval & RAGRanking & reranking | |
| XERJ systemone gategithub.com | — | XERJ, a local AI search engine, exposes a Jev-compatible /v1/systemone endpoint so the off-the-shelf jev-reranker client reranks a search index against it unmodified. | Search, retrieval & RAGRanking & reranking | |
| XERJ systemone gate — appxerj.org | — | XERJ, a local AI search engine, exposes a Jev-compatible /v1/systemone endpoint so the off-the-shelf jev-reranker client reranks a search index against it unmodified. | Search, retrieval & RAGRanking & reranking |
Semantic search22 of 22 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| AILIKE for MySQLgithub.com | 4 | Native MySQL plugin that adds an AILIKE operator and ailike() functions for filtering rows and comparing text columns with natural-language conditions, each judged by Jev. | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 2 | Bentlybro/siftr — Fast, cheap judgment for AI coding agents: semantic search, focused reads and list picking in ~2s. CLI + MCP server on TypeSafe Jev. Benchmarked on SWE-bench. · endpoint · Python | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 5 | Peu77/JevFind — Fast semantic code search powered by Jev. Find the relevant files, line ranges, and snippets · endpoint · Rust | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 75 | can1357/jegrep — Semantic grep: find code by describing what you're looking for, powered by Jev. · endpoint · Rust | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 1 | kazuhideoki/jev-search — Recursive semantic file search using TypeSafe Jev and fzf · endpoint · JavaScript | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 131 | keltokhy/jgrep — grep, but the pattern is a description. Filters lines by meaning with TypeSafe's Jev decision model: ~200 ms and a thousandth of a cent per line. · endpoint · Python | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 59 | kyu1204/jgrep — grep for what code does, not what it's called. Semantic code search powered by TypeSafe Jev. · endpoint · TypeScript | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 6 | larguesa/jev-search — Experimental semantic line search with TypeSafe Jev via OpenRouter. Python CLI with no runtime dependencies. · endpoint · Python | Search, retrieval & RAGSemantic search | |
| endpointgithub.com | 145 | uehaj/jev-semgrep — grep by meaning, across languages. TypeSafe Jev scores every line against a meaning; combine meanings with AND/OR/NOT. 意味で探す grep。日本語で英語を、英語で日本語を検索できる · endpoint · JavaScript | Search, retrieval & RAGSemantic search | |
| Intent-aware search in Replicasx.com | — | Experiment powering the global search in the Replicas app with Jev so results and actions follow what the user means. | Search, retrieval & RAGSemantic search | |
| Jev Code Findergithub.com | 5 | Rust CLI and agent skill for semantic code search: describe a concept in plain English and get relevant files, line ranges, confidence scores, and snippets. | Search, retrieval & RAGSemantic search | |
| jev-searchgithub.com | 6 | Dependency-free Python CLI for semantic line search over documents, notes, and knowledge bases via Jev on OpenRouter, meant to complement exact keyword search for agents. | Search, retrieval & RAGSemantic search | |
| jev-semgrepgithub.com | 145 | Greps by meaning instead of by regex: every line gets a probability from Jev, and meanings combine with AND, OR, and NOT. | Search, retrieval & RAGSemantic search | |
| JevSQLgithub.com | 5 | Node library and CLI that adds natural-language predicates to SQLite, letting queries filter, rank, classify, and score rows by meaning with batching, caching, cost guards, and review queues. | Search, retrieval & RAGSemantic search | |
| kazuhideoki/jev-searchgithub.com | 1 | Recursive semantic file search using TypeSafe Jev and fzf | Search, retrieval & RAGSemantic search | |
| model_idgithub.com | 1 | zaydmulani09/jevgrep — Instant semantic code search: local ripgrep shortlist, meaning-ranked by TypeSafe Jev via OpenRouter. · model_id · Python | Search, retrieval & RAGSemantic search | |
| Needlewww.theunwindai.com | — | Open-source Chrome extension for searching by meaning instead of words: ask a question and Jev scores the page's sentences and highlights the ones that match your intent, with no generated answer. | Search, retrieval & RAGSemantic search | |
| Semantic Find (⌘F) extensionx.com | — | Open-source Chrome extension that replaces Find on page with semantic matching, highlighting passages that mean what you typed in near real time. | Search, retrieval & RAGSemantic search | |
| Semantic object filterx.com | — | Small interactive demo where you type a phrase such as things you can wear in winter and Jev picks out the matching objects on screen. | Search, retrieval & RAGSemantic search | |
| Senseekgithub.com | 0 | Browser extension that searches the page you are reading by meaning from a Ctrl+F-style box and jumps to ranked passages, using your own Jev API key with no backend. | Search, retrieval & RAGSemantic search | |
| YC startup semantic searchx.com | — | Search engine over 6000+ Y Combinator startups that answers free-form queries (color, niche, competitor, age, image) in under a second, built with Jev for $2.7 in total testing costs. | Search, retrieval & RAGSemantic search | |
| zaydmulani09/jevgrepgithub.com | 1 | Instant semantic code search: local ripgrep shortlist, meaning-ranked by TypeSafe Jev via OpenRouter. | Search, retrieval & RAGSemantic search |
Web research19 of 19 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| deep-searchergithub.com | — | Data & Search: Open Source Deep Research alternative with Jev search-stopping evaluation. | Search, retrieval & RAGWeb research | |
| endpointgithub.com | 495 | superagents-lab/jev-search — Search the web with TypeSafe's Jev: source selection, query understanding and relevance ranking. Built with Search1API. · endpoint · TypeScript | Search, retrieval & RAGWeb research | |
| Jev Radargithub.com | — | Local research workspace where Jev steers an investigation over public sources, asking the same research questions across records to build an evidence-linked comparison with every decision inspectable. | Search, retrieval & RAGWeb research | |
| jev-research-pipelinegithub.com | 3 | Daily research monitor for standing questions: deterministic Python owns the loop, TypeSafe Jev screens sources per question, Qwen writes the notes (pilot) | Search, retrieval & RAGWeb research | |
| jev-scoutgithub.com | — | MCP server for agent web research that has Jev judge every query, search result, and fetched page for relevance and credibility, with session budgets, SSRF-guarded fetching, and a live decision dashboard. | Search, retrieval & RAGWeb research | |
| jev-scout — studygithub.com | — | MCP server for agent web research that has Jev judge every query, search result, and fetched page for relevance and credibility, with session budgets, SSRF-guarded fetching, and a live decision dashboard. | Search, retrieval & RAGWeb research | |
| jev-searchgithub.com | 59 | Claude Code plugin in the muki-ai-plugins marketplace that answers WebSearch with Jev-ranked results, where Jev chooses sources and time window, falling back to the built-in tool. | Search, retrieval & RAGWeb research | |
| jev-searchgithub.com | 495 | Natural-language web search that picks the time range and query with Choices and keeps each source only if its Noul passes (needs a Search1API key). | Search, retrieval & RAGWeb research | |
| jev-search — demox.com | — | Natural-language web search that picks the time range and query with Choices and keeps each source only if its Noul passes (needs a Search1API key). | Search, retrieval & RAGWeb research | |
| jev-search — marketplacegithub.com | — | Claude Code plugin in the muki-ai-plugins marketplace that answers WebSearch with Jev-ranked results, where Jev chooses sources and time window, falling back to the built-in tool. | Search, retrieval & RAGWeb research | |
| SearchBoost Jev evidence loopgithub.com | — | Multi-engine web search and evidence synthesis for coding agents (MCP server, Pi extension, DeepSeek Harness bundle) with an experimental adaptive_search loop that uses Jev to research targets in batches. | Search, retrieval & RAGWeb research | |
| SearchBoost Jev evidence loop — repogithub.com | — | Multi-engine web search and evidence synthesis for coding agents (MCP server, Pi extension, DeepSeek Harness bundle) with an experimental adaptive_search loop that uses Jev to research targets in batches. | Search, retrieval & RAGWeb research | |
| SlopSearXgithub.com | 9 | Stateless meta search engine for AI agents and drop-in SearXNG replacement that asks Jev which specialist engines can add distinctive evidence for each query. | Search, retrieval & RAGWeb research | |
| SlopSearX — docsgithub.com | 9 | Stateless meta search engine for AI agents and drop-in SearXNG replacement that asks Jev which specialist engines can add distinctive evidence for each query. | Search, retrieval & RAGWeb research | |
| Smart Search Jevgithub.com | — | Smart Search, a desktop app that pulls current web information into AI conversations, uses Jev for typed judgments while keeping execution and the allowed action set in code. | Search, retrieval & RAGWeb research | |
| Web Search Plus Jevgithub.com | 414 | Web Search Plus, a Hermes agent plugin for multi-provider search and page extraction, adds optional Jev decisions at gated seams, off by default and confidence-thresholded. | Search, retrieval & RAGWeb research | |
| web-search-plus-mcpgithub.com | 5 | MCP server that gives agents web search across 15 search and 9 extract providers with original sources; optional, off-by-default Jev confirms news-type searches, scores extracted bodies and fills in missing language. | Search, retrieval & RAGWeb research | |
| web-search-plus-mcp — appwebsearchplus.xyz | — | MCP server that gives agents web search across 15 search and 9 extract providers with original sources; optional, off-by-default Jev confirms news-type searches, scores extracted bodies and fills in missing language. | Search, retrieval & RAGWeb research | |
| webctl — demox.com | — | Web search CLI for agents that queries three search backends, has Jev score each result set against the query and goal, and dedupes the high-scoring subset so the agent reads far less. | Search, retrieval & RAGWeb research |
Knowledge graphs1 of 1 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| jevgraphgithub.com | 32 | Parses PDF, DOCX, PPTX, or text locally, then asks Jev one closed-set relation question per candidate entity pair and exports a graph with per-edge probabilities and page evidence to JSON, CSV, or Neo4j. | Search, retrieval & RAGKnowledge graphs |
More projects & source code67 of 67 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| ai_sdkgithub.com | 6 | Charlyhno-eng/jev-document-classification — JEV Document Classification enables the rapid and cost-effective classification of text-based documents using AI, leveraging TypeSafe's "System One" model. · ai_sdk · TypeScript | Search, retrieval & RAGMore projects & source code | |
| avgithub.com | 7 | Video memory CLI for agents that indexes captions and transcripts for search and Q&A; with a TypeSafe key, av ask has Jev filter and rank retrieved scenes and check whether the answer is supported. | Search, retrieval & RAGMore projects & source code | |
| Book Learning Jev retrieval gategithub.com | — | Retrieval gate in the self-hosted BOOKNOTE textbook RAG study assistant: Jev scores each fused candidate chunk for relevance and screens it for prompt injection before it reaches the answer, failing open to the ungated results. | Search, retrieval & RAGMore projects & source code | |
| Book Learning Jev retrieval gate — repogithub.com | — | Retrieval gate in the self-hosted BOOKNOTE textbook RAG study assistant: Jev scores each fused candidate chunk for relevance and screens it for prompt injection before it reaches the answer, failing open to the ungated results. | Search, retrieval & RAGMore projects & source code | |
| bootcampgithub.com | — | Data & Search: Runnable search tutorials using Gemini embeddings, Milvus retrieval, and Jev judgments for reranking, filtering, and routing. | Search, retrieval & RAGMore projects & source code | |
| cascade-searchgithub.com | 8 | In-browser search query parser: a 27,193-parameter model labels every word with a role and calibrated confidence in about 0.25 ms, and only the uncertain words are escalated to Jev. | Search, retrieval & RAGMore projects & source code | |
| DataJevgithub.com | 1 | Data-analysis agent where an LLM writes each Python analysis step and Jev reads a compressed state of goal, schema, recent results and insights to decide whether to continue, switch direction, verify or stop. | Search, retrieval & RAGMore projects & source code | |
| Dodo Jev web toolsgithub.com | — | Coding agent on Cloudflare Workers whose read-only web tools use Jev to rank candidate URLs before fetching and to check whether a fetched page answers the query. | Search, retrieval & RAGMore projects & source code | |
| Dodo Jev web tools — repogithub.com | — | Coding agent on Cloudflare Workers whose read-only web tools use Jev to rank candidate URLs before fetching and to check whether a fetched page answers the query. | Search, retrieval & RAGMore projects & source code | |
| endpointgithub.com | 5 | EugeneBoondock/jevsql — SQL with natural-language predicates, powered by TypeSafe's Jev. Filter, rank, classify and score rows by meaning — batched, cached and cost-guarded. · endpoint · JavaScript | Search, retrieval & RAGMore projects & source code | |
| endpointgithub.com | 3 | baronunread/leanest — Local-first test selector using Jev judgments to determine which tests are affected by a code change · endpoint · TypeScript | Search, retrieval & RAGMore projects & source code | |
| endpointgithub.com | 36 | devanshbatham/commit-miner — Classify Git commit diffs and messages with Jev. Bug fixes, security fixes/CWEs, and change types. · endpoint · Rust | Search, retrieval & RAGMore projects & source code | |
| endpointgithub.com | 1 | ibnuh/Flow.Launcher.Plugin.JevFileSearch — Flow Launcher predictive file search with Jev intent reranking. Type natural language like 'the pdf I just downloaded' and get the right file. · endpoint · C# | Search, retrieval & RAGMore projects & source code | |
| endpointgithub.com | 485 | kyotofin/tax-doc-classifier — Tax document page classifier built on Jev decisions. 100% strict accuracy across 261 IRS forms, ~$0.001 per page. · endpoint · TypeScript | Search, retrieval & RAGMore projects & source code | |
| endpointgithub.com | 4 | maayanlevy/mysql-ailike — Natural-language row filtering for MySQL, powered by TypeSafe Jev. · endpoint · C++ | Search, retrieval & RAGMore projects & source code | |
| endpointgithub.com | 7 | sufianetaouil/every — Ask a yes/no question of every function in a codebase. Ranked answers in seconds, for cents. Grep whose pattern is a question, powered by TypeSafe Jev. · endpoint · Python | Search, retrieval & RAGMore projects & source code | |
| enskgithub.com | — | English-Icelandic dictionary on Cloudflare Workers that combines D1 full-text search and Vectorize semantic recall, then has Jev via Workers AI rerank the shortlist and decide whether any entry truly matches. | Search, retrieval & RAGMore projects & source code | |
| Enzymegithub.com | 82 | Local-first compile step for Markdown knowledge bases whose enzyme compile uses Jev via OpenRouter Decisions to scan the vault and write the program that decides where catalyst questions are learned from. | Search, retrieval & RAGMore projects & source code | |
| etnt/unsafe-c-findergithub.com | 1 | Classify C/C++ snippets via TypeSafe Jev through OpenRouter. | Search, retrieval & RAGMore projects & source code | |
| findmegithub.com | 9 | Rust CLI that finds a file or folder from a natural-language memory of it by walking the directory tree and asking Jev which entries at each level look most promising. | Search, retrieval & RAGMore projects & source code | |
| FindSFSymbolsgithub.com | 5 | Mac app for finding SF Symbols by plain description: on-device word vectors shortlist 48 symbols, then Jev answers 48 yes/no questions in one request and the scores drive a physics pile where good matches float up. | Search, retrieval & RAGMore projects & source code | |
| GPTCachegithub.com | — | Data & Search: Semantic cache with a Jev evaluator that uses Noul judgments to check whether a cached response can serve an incoming request. | Search, retrieval & RAGMore projects & source code | |
| Hacker News For Megithub.com | 1 | Personal Hacker News reader that screens new stories against your saved interests and shows only the relevant ones in an HN-style feed. | Search, retrieval & RAGMore projects & source code | |
| ibnuh/Flow.Launcher.Plugin.JevFileSearchgithub.com | 1 | Flow Launcher predictive file search with Jev intent reranking. Type natural language like 'the pdf I just downloaded' and get the right file. | Search, retrieval & RAGMore projects & source code | |
| Jev and SerpApi fact checkergithub.com | — | Tutorial CLI from SerpApi that checks a statement or yes/no question against up to five Google organic snippets and gets a verdict from Jev through OpenRouter's Decisions API. | Search, retrieval & RAGMore projects & source code | |
| Jev and SerpApi fact checker — repogithub.com | — | Tutorial CLI from SerpApi that checks a statement or yes/no question against up to five Google organic snippets and gets a verdict from Jev through OpenRouter's Decisions API. | Search, retrieval & RAGMore projects & source code | |
| Jev Papersgithub.com | 3 | Classifies 1,000 recent arXiv AI papers into 24 topics with one Jev decision each, checks a 100-paper sample against Claude Opus 5, and renders a searchable topic map. | Search, retrieval & RAGMore projects & source code | |
| Jev Search for Obsidiangithub.com | — | Obsidian plugin that searches the vault offline with BM25 and then, only after you approve what gets sent, reranks the top results with Jev. | Search, retrieval & RAGMore projects & source code | |
| jev-bigquery-cloudrungithub.com | — | Data & Search: Classify support tickets in BigQuery with Jev and Cloud Run | Search, retrieval & RAGMore projects & source code | |
| jev-folio-recursive-classifiergithub.com | 1 | Classifier that walks OCR'd legal agreements down the FOLIO Document Types ontology with one Jev Choice per level, a beam of candidate paths, and confidence-gated stopping at leaves. | Search, retrieval & RAGMore projects & source code | |
| jev-harnessgithub.com | — | Data & Search: A Pi agent harness built around TypeSafe's Jev (System One model): router, context picker, gate, verifier. Tested on Neon Postgres branches. | Search, retrieval & RAGMore projects & source code | |
| jev-information-extractiongithub.com | 2 | Upload a PDF and ask questions such as the GST number or invoice total; Jev ranks the extracted text chunks that answer each one and the app shows each match, its probability and its location on the page. | Search, retrieval & RAGMore projects & source code | |
| jev-retrievalgithub.com | — | Data & Search: Grep-shaped Rust CLI for coding agents that finds files matching a plain-language concept: a local BM25 pass recalls candidates, Jev verifies each file window-by-window with Noul gates, and one listwise Choice per lane reranks the kept files into \path:line score\ output with calibrated probabilities. (Author-submitted: I am the maintainer.) | Search, retrieval & RAGMore projects & source code | |
| jev-reviewsgithub.com | — | Data & Search: The flagship experiment in this workspace is \\\jev-reviews\\\: an automated pipeline that ingests Google Maps restaurant reviews from \\Apify Storage\\, stores and deduplicates them in \\Supabase PostgreSQL\\, and classifies them using \\Jev (TypeSafe System One)\\ or traditional generative LLMs across 5 operational pillars, extracting dishes, customer highlights, and owner recommendations. | Search, retrieval & RAGMore projects & source code | |
| jev-upwork-job-classificationgithub.com | 2 | Upwork job classification example | Search, retrieval & RAGMore projects & source code | |
| JevDeepResearchgithub.com | — | Data & Search: GPT directs the research; Jev uses Choice and Noul to locate evidence across document regions concurrently, and code returns original passages for GPT to verify and continue. The released integration pairs Pi-Serini BM25 search with Jev batches of 20, 40, or 60 documents. | Search, retrieval & RAGMore projects & source code | |
| jeverywordgithub.com | 5 | Dependency-free JavaScript library for field extraction, PII detection, and exact quotes that numbers the words of a text so Jev can pick them, returning original substrings with offsets and probabilities. | Search, retrieval & RAGMore projects & source code | |
| Jevflixgithub.com | 1 | Jev picks, you watch. A hybrid movie recommender: fast semantic + keyword search narrows 4,800 films to a shortlist, then TypeSafe Jev reads your constraints and picks the one film that fits - with a confidence score that decides whether to answer instantly or ask a follow-up. | Search, retrieval & RAGMore projects & source code | |
| jevframegithub.com | 20 | Python library for pandas and Polars that labels, scores, and classifies DataFrame rows with natural-language questions, returning full probability distributions as ordinary Series with bounded async concurrency. | Search, retrieval & RAGMore projects & source code | |
| JevPDFgithub.com | 5 | Searches a PDF by meaning in the browser: pdf.js extracts the lines locally, Jev answers one Noul per line in batches of 16, and matching lines light up page by page, ranked by probability. | Search, retrieval & RAGMore projects & source code | |
| jevsqlgithub.com | — | Data & Search: Text-to-SQL where the model never writes SQL — typed, calibrated decisions (TypeSafe Jev) + code-assembled queries | Search, retrieval & RAGMore projects & source code | |
| jevtok-tsgithub.com | — | Data & Search: \\No Python, network calls, API keys, native extensions, or runtime dependencies.\\ Vocabulary and lookup tables ship inside the package. Ordinary \o200k\_base\ tokenization is not interchangeable with Jev: \evidence\ is three tokens and \2024\ is four. | Search, retrieval & RAGMore projects & source code | |
| Job Seek Jev AI filtergithub.com | — | Open-source job search built from 5,300+ career sites that adds a Jev AI filter classifying each posting as accepted or rejected against a user's criteria. | Search, retrieval & RAGMore projects & source code | |
| longjevgithub.com | 1 | Experimental wrapper with the same system_one(state, questions) call as Jev that accepts inputs beyond Jev's 32K-token limit by scoring every chunk, keeping the best ones and asking Jev over what remains. | Search, retrieval & RAGMore projects & source code | |
| Milvus Search with Jevgithub.com | — | Nine runnable notebooks combining Gemini embeddings and Milvus retrieval with Jev decisions for reranking, filtering, search stopping, routing, cache reuse, curation, guardrails, and evaluation. | Search, retrieval & RAGMore projects & source code | |
| model_idgithub.com | 4 | 1jehuang/jev-pr-labeler — Semantic GitHub PR labels using Jev's typed decisions, with conceptual scope instead of line counts · model_id · Python | Search, retrieval & RAGMore projects & source code | |
| model_idgithub.com | 1 | etnt/unsafe-c-finder — Classify C/C++ snippets via TypeSafe Jev through OpenRouter. · model_id · Python | Search, retrieval & RAGMore projects & source code | |
| NYC 311 JEV Heatmapsgithub.com | 2 | Reproducible pipeline that evaluates NYC 311 complaint descriptions and maps the resulting impact signals on a geographic grid as interactive HTML maps and PNGs; the live run made 634 API calls. | Search, retrieval & RAGMore projects & source code | |
| obsidian-tc Jev citation judgegithub.com | — | Opt-in Jev judge provider in obsidian-tc, a governed Obsidian MCP server with fused retrieval and vault memory, used only to decide whether inferred citations between notes hold up. | Search, retrieval & RAGMore projects & source code | |
| obsidian-tc Jev citation judge — repogithub.com | — | Opt-in Jev judge provider in obsidian-tc, a governed Obsidian MCP server with fused retrieval and vault memory, used only to decide whether inferred citations between notes hold up. | Search, retrieval & RAGMore projects & source code | |
| Patterns of Design searchgithub.com | — | Search on the patternsof.design pattern-language site where Jev scores how much practical guidance each design pattern gives for what the user wants to build, directly or by analogy. | Search, retrieval & RAGMore projects & source code | |
| Patterns of Design search — repogithub.com | — | Search on the patternsof.design pattern-language site where Jev scores how much practical guidance each design pattern gives for what the user wants to build, directly or by analogy. | Search, retrieval & RAGMore projects & source code | |
| Polar Llama TypeSafe support — repogithub.com | — | Native Rust TypeSafe layer in the Polars LLM plugin that answers Noul, Choice, and Score questions per row, or applies a Pydantic contract to every line of a document, as typed dataframe columns. | Search, retrieval & RAGMore projects & source code | |
| RAGElo Jev evaluatorsgithub.com | 131 | Elo-based toolkit for evaluating RAG agents that can use Jev via TypeSafe or Vercel AI Gateway as a retrieval relevance judge and pairwise answer evaluator. | Search, retrieval & RAGMore projects & source code | |
| Research Enginegithub.com | 5 | Evidence-first research runtime for AI agents that routes a question to research packs and read-only connectors; an optional jev-triage command asks Jev how relevant already-collected public evidence rows are. | Search, retrieval & RAGMore projects & source code | |
| sdkgithub.com | 2 | ekkyarmandi/jev-upwork-job-classification — Upwork job classification example · sdk · Python | Search, retrieval & RAGMore projects & source code | |
| sdkgithub.com | 92 | ellipsis-dev/blink — Codebase search powered by Jev from @typesafe-ai · sdk · TypeScript | Search, retrieval & RAGMore projects & source code | |
| Search-Function-Testgithub.com | 0 | A test project based on Jev AI, the goal is to build a search function for a blog/article website that has 100s of articles to search from, So the user can actually use the search as chat to question anything and find related answers/articles | Search, retrieval & RAGMore projects & source code | |
| second-brain Jev classifiergithub.com | — | Ingestion classifier in second-brain, an LLM-curated private search engine over Slack, GitHub, Drive and personal messages, that asks Jev to assign each SMS, Gmail or call transcript a segment and a retention score. | Search, retrieval & RAGMore projects & source code | |
| second-brain Jev classifier — repogithub.com | — | Ingestion classifier in second-brain, an LLM-curated private search engine over Slack, GitHub, Drive and personal messages, that asks Jev to assign each SMS, Gmail or call transcript a segment and a retention score. | Search, retrieval & RAGMore projects & source code | |
| sievegithub.com | 1 | Sift a whole folder of notes by any judgment powered by Jev at OpenRouter | Search, retrieval & RAGMore projects & source code | |
| Starguidegithub.com | 16 | Serverpod's documentation chatbot where Jev picks the doc and website pages that answer a question, judges whether they contain the answer so vector search can be skipped, and rates whether the final answer resolved it. | Search, retrieval & RAGMore projects & source code | |
| tax-doc-classifiergithub.com | 485 | One request per page picks among 261 IRS forms and seven page kinds; reports 100 percent on its corpus at $0.001 a page, 34 times cheaper than the LLM pipeline it replaced. | Search, retrieval & RAGMore projects & source code | |
| Transcript Lensgithub.com | 16 | Next.js app with a Turkish interface for exploring YouTube transcripts by meaning, where Jev classifies each block without rewriting it to surface chapters, key passages, and topic mentions. | Search, retrieval & RAGMore projects & source code | |
| vector-graph-raggithub.com | — | Data & Search: Graph RAG with pure vector search, achieving SOTA performance in multi-hop reasoning scenarios. | Search, retrieval & RAGMore projects & source code | |
| Wordcellgithub.com | 14 | Local Markdown knowledge base for coding agents with an optional Jev reranking lane; on SciFact it put a relevant result first for 161 of 300 queries versus 101 with exact search alone. | Search, retrieval & RAGMore projects & source code | |
| Wordcell — docsgithub.com | 14 | Local Markdown knowledge base for coding agents with an optional Jev reranking lane; on SciFact it put a relevant result first for 161 of 300 queries versus 101 with exact search alone. | Search, retrieval & RAGMore projects & source code |
More posts & discussions16 of 16 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| Cascade Searchx.com | — | In-browser search bar where a 27K-parameter model parses queries like "open bugs from sam since last week" into typed filters in 0.25 ms and asks Jev only about words it is unsure of. | Search, retrieval & RAGMore posts & discussions | |
| Embedded databases plus Jev rerankingx.com | — | Architecture write-up pairing SQLite, DuckDB and LadyBugDB for hybrid candidate retrieval with Jev as the reranker, aiming for sub-second RAG results instead of sending 20 to 50 passages to a frontier LLM. | Search, retrieval & RAGMore posts & discussions | |
| Fuzzy content searchx.com | — | Search site over the author's own published content where Jev scores each item against a vague query and shows the scores, which makes the ranking easy to inspect. | Search, retrieval & RAGMore posts & discussions | |
| Hacker News For Me — demonews.ycombinator.com | — | Personal Hacker News reader that screens new stories against your saved interests and shows only the relevant ones in an HN-style feed. | Search, retrieval & RAGMore posts & discussions | |
| Hacker News For Me — demo 2x.com | — | Personal Hacker News reader that screens new stories against your saved interests and shows only the relevant ones in an HN-style feed. | Search, retrieval & RAGMore posts & discussions | |
| Hacker News For Me — discussionwww.reddit.com | — | Personal Hacker News reader that screens new stories against your saved interests and shows only the relevant ones in an HN-style feed. | Search, retrieval & RAGMore posts & discussions | |
| Hacker News personal recommenderx.com | — | Experiment that has Jev pick which Hacker News front-page stories match the author's interests each morning, with the composite-judgment design written up in a gist. | Search, retrieval & RAGMore posts & discussions | |
| Hazumi Newsnews.ycombinator.com | — | Hacker News reader that uses Jev to filter large discussion threads down to the comments worth reading. | Search, retrieval & RAGMore posts & discussions | |
| Marginx.com | — | Tool that imports years of X bookmarks and lets you search them in plain language, with Jev scoring every saved tweet in parallel and returning a relevance-ranked list in seconds. | Search, retrieval & RAGMore posts & discussions | |
| Margin — demox.com | — | Tool that imports years of X bookmarks and lets you search them in plain language, with Jev scoring every saved tweet in parallel and returning a relevance-ranked list in seconds. | Search, retrieval & RAGMore posts & discussions | |
| Paper Breakdown recommendationsx.com | — | Adds Jev curation on top of the content-based and collaborative-filtering recommender in Paper Breakdown, at $0.0019 per user for ~70 recs. | Search, retrieval & RAGMore posts & discussions | |
| Physical library index searchx.com | — | Searches the author's physical books by checking an entire book index in parallel in 300ms, e.g. pointing "Who invented photography?" to page 98 under Niepce. | Search, retrieval & RAGMore posts & discussions | |
| Sanity content agent without LLM queriesx.com | — | Demo pairing Jev with the Sanity CMS where answers are pre-written at ingest, so the agent selects an answer, skips retrieval when it can and responds in under a second instead of 4+. | Search, retrieval & RAGMore posts & discussions | |
| SF food permit mapx.com | — | Map of permit and health-inspection records for every food and beverage business in San Francisco, where Jev answers questions about the records in seconds. | Search, retrieval & RAGMore posts & discussions | |
| Tripwire internet scannerwww.reddit.com | — | Early-access tool that scans Reddit, X, and LinkedIn posts and filters them by concepts defined in natural language instead of keywords. | Search, retrieval & RAGMore posts & discussions | |
| Zillow natural-language searchx.com | — | Scans thousands of Zillow listings and classifies them by things the site has no filter for, like architecture, renovation status or distance to freeways, in under 20 seconds for $0.18. | Search, retrieval & RAGMore posts & discussions |
More guides & websites15 of 15 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| av — appagentic.video | — | Video memory CLI for agents that indexes captions and transcripts for search and Q&A; with a TypeSafe key, av ask has Jev filter and rank retrieved scenes and check whether the answer is supported. | Search, retrieval & RAGMore guides & websites | |
| Cascade Search — appcascade.scira.ai | — | In-browser search bar where a 27K-parameter model parses queries like "open bugs from sam since last week" into typed filters in 0.25 ms and asks Jev only about words it is unsure of. | Search, retrieval & RAGMore guides & websites | |
| Enzyme — appmemory.enzyme.garden | — | Local-first compile step for Markdown knowledge bases whose enzyme compile uses Jev via OpenRouter Decisions to scan the vault and write the program that decides where catalyst questions are learned from. | Search, retrieval & RAGMore guides & websites | |
| FindSFSymbols — apptornikegomareli.github.io | — | Mac app for finding SF Symbols by plain description: on-device word vectors shortlist 48 symbols, then Jev answers 48 yes/no questions in one request and the scores drive a physics pile where good matches float up. | Search, retrieval & RAGMore guides & websites | |
| Fuzzy content search — appgigabit-search.gigabitmillion-games.workers.dev | — | Search site over the author's own published content where Jev scores each item against a vague query and shows the scores, which makes the ranking easy to inspect. | Search, retrieval & RAGMore guides & websites | |
| Hacker News personal recommender — gistgist.github.com | — | Experiment that has Jev pick which Hacker News front-page stories match the author's interests each morning, with the composite-judgment design written up in a gist. | Search, retrieval & RAGMore guides & websites | |
| Hazumi News — appwww.hazumi.news | — | Hacker News reader that uses Jev to filter large discussion threads down to the comments worth reading. | Search, retrieval & RAGMore guides & websites | |
| Hazumi News — projecthazumi.news | — | Hacker News reader that uses Jev to filter large discussion threads down to the comments worth reading. | Search, retrieval & RAGMore guides & websites | |
| jev-information-extraction — appjev-information-extraction-fibby-prod-telegram.up.railway.app | — | Upload a PDF and ask questions such as the GST number or invoice total; Jev ranks the extracted text chunks that answer each one and the app shows each match, its probability and its location on the page. | Search, retrieval & RAGMore guides & websites | |
| Job Seek Jev AI filter — appjseek.co | — | Open-source job search built from 5,300+ career sites that adds a Jev AI filter classifying each posting as accepted or rejected against a user's criteria. | Search, retrieval & RAGMore guides & websites | |
| Patterns of Design search — apppatternsof.design | — | Search on the patternsof.design pattern-language site where Jev scores how much practical guidance each design pattern gives for what the user wants to build, directly or by analogy. | Search, retrieval & RAGMore guides & websites | |
| Search-Function-Testsearch-function-test.vercel.app | — | A test project based on Jev AI, the goal is to build a search function for a blog/article website that has 100s of articles to search from, So the user can actually use the search as chat to question anything and find related answers/articles | Search, retrieval & RAGMore guides & websites | |
| Starguide — packagepub.dev | — | Serverpod's documentation chatbot where Jev picks the doc and website pages that answer a question, judges whether they contain the answer so vector search can be skipped, and rates whether the final answer resolved it. | Search, retrieval & RAGMore guides & websites | |
| Tripwire internet scanner — apptripwire.easytech-agency.net | — | Early-access tool that scans Reddit, X, and LinkedIn posts and filters them by concepts defined in natural language instead of keywords. | Search, retrieval & RAGMore guides & websites | |
| Wordcell — appwordcell.io | — | Local Markdown knowledge base for coding agents with an optional Jev reranking lane; on SciFact it put a relevant result first for 161 of 300 queries versus 101 with exact search alone. | Search, retrieval & RAGMore guides & websites |
No matches this time
Try a broader search or reset your filters to explore the full collection.
Showing 193 of 193 resources