Live radar · scored by Jev

What people are building with Jev.

119 signals · last crawl 1h ago

Filter

Jev Skill Review

Jev is a decision model used to review a corpus of skills, providing ratings across rubric dimensions and identifying skills that may need distillation.

SCORE

@Kunal_Jain93h

Jev: Rule Selection for Claude Code

Jev intelligently selects applicable rules for Claude Code based on prompts and file paths, optimizing context token usage and ensuring relevant rule delivery.

FILTER

@EliaAlberti1h

Jev for Code Reviews

Jev is a tool that scores raw git diffs directly to identify issues, invoking a language model only when an issue is flagged, significantly improving speed and reducing cost for code reviews.

SCORE

@liorshkiller1h

Shipped sqlite-jev: loadable sqlite extension + Python wrapper for asking @typesafeai #Jev typed

Noul, Choice, and Score via scalar functions and batched virtual-table queries, with confidence attached. https://github.com/mgaitan/sqlite-jev https://x.com/tin_nqn_/status/2101030999945580598/photo/1

CHOOSE

@tin_nqn_1h

Jev Integration with Werkfaden

Jev is a new component integrated into Werkfaden that provides typed probabilistic decisions about project state, rather than generating text.

VERIFY

@yawende861h

Tenet: AI Code Review Gate

Tenet is a review gate for code generated by agents, evaluating commits, commit messages, and pull requests against plain-language rules using Jev.

GATE

@timche_1h

Agent 'Done' Claim Verification

This project verifies agent claims of task completion using Jev to prevent false 'done' signals.

GATE

@dyfrai1h

Jev Watch

Jev Watch is an open-source tool designed to monitor Jev model updates by committing correct answers and re-running them after updates to flag any drift.

VERIFY

@akshaykanthed4h

DLQ Inspector with Jev Integration

DLQ Inspector analyzes and classifies failed messages from dead-letter queues, now optionally using Jev by TypeSafe AI for classification when existing rule-based systems are insufficient.

CHOOSE

@HalxDocs3h

Kev-0.5B: A Tiny Open-Source Jev-like Decision Model

Kev-0.5B is a small, open-source decision model that utilizes LoRA and a readout head on Qwen2.5-0.5B, featuring a TypeSafe-compatible API for parallel typed questions with learned probabilities, and operates without decoding.

CHOOSE

@shipfrontierai3h

JEV-SEO

A Rust CLI tool that audits web pages for SEO and checks AI crawler accessibility using a typesafe-ai JEV model.

VERIFY

@Akash__ydv0014h

TypeSafe Jev Model and Gemini Integration Test

This project tested the TypeSafe Jev model alongside Gemini across 10 development stages, observing significant reductions in token footprint, latency, and evaluation cost.

GATE

@IranUsher1h

Jev-Powered Code Review and CI/CD Tooling Enhancement

This project enhances code review and CI/CD tooling by optimizing task preparation, context selection, and model routing to match routine work with cheaper models and reserve deeper review for critical changes, ensuring quality is measurable through independent checks and recorded outcomes.

ROUTE

@sam_builds_ai2h

Jevis: Natural Language Integration Tests for Flutter

Jevis is a Flutter package that enables writing integration tests for Flutter applications using natural language, powered by Jev from TypeSafe.ai.

VERIFY

@jake_gwon3h

JevQL

JevQL provides semantic WHERE clauses for vanilla Postgres, powered by Jev.

FILTER

@kylemclaren1h

Jev Docs Confusion Heatmap

This project uses Jev to analyze technical documentation, identifying sections that may be difficult for beginners by detecting unexplained terminology, assumed knowledge, unclear relationships, ambiguous instructions, and information density, then visualizing these as a confusion heatmap.

SCORE

@riseandshaheen3h

Jev CLI Tool

An unofficial open-source command-line interface and stdio MCP server for Jev, supporting multiple APIs and offering JSON output for scripting.

CHOOSE

@DevTumf1h

Jev-Enhanced Tool Calling for Small Models

This project improves a small 0.6B model's tool-calling capabilities by using Jev to select the best response from multiple attempts and training on those selections.

CHOOSE

@hsingh_txt3h

Pi Jev Context Curator

This project provides a smaller context for model calls by semantically pruning context with Jev.

FILTER

@ShashankH_1h

Jev Build Gating

Jev, an independent index by @typesafeai, now gates Tryeve's builds by performing checks before a full sandbox run, providing instant feedback at zero cost.

GATE

@varde_abhi1h

OpenClaw Typesafe AI Plugin

This project adds an optional OpenClaw tool for typed Jev decisions, allowing users to input state and questions to receive answers from a defined space.

CHOOSE

@Olli01031h

Jev Typed Question Answering Model

Jev is a novel model that answers typed questions with associated probabilities, operating in approximately half a second.

SCORE

@mihai_respira1h

Jev MCP Tool

This project built a Jev MCP Tool that saves time and tokens on ranking tasks for LLM models, but requires explicit instructions on how to process the Jev data.

RANK

@unbug3h

Claude Code Plugin for Jev

A Claude Code plugin that scores how well users prompt their coding agent, with no added latency.

SCORE

@prateekkathal1h