Jev ecosystem projects — page 126

No project description available.

Experiments with TypeSafe Jev + Browser Use (based on browser-use/jev-ultrafast, MIT)

MCP tool exposing TypeSafe's Jev browser agent (via browser-use/jev-ultrafast) to Claude Code/Desktop

Jev-powered browser automation for Codex and Claude: native reasoning, continuous action loops, CLI, MCP and Chrome extension.

Fast browser QA from Claude Code or pi: TypeSafe Jev picks every click via Jev Ultrafast, you supply text and the pass check.

Exact-target audit of the probabilities returned by calibration-trained decision models (TypeSafe Jev, RLCD)

Jev(TypeSafe)가 요청을 먼저 판정하고 필요한 작업만 알맞은 Claude 모델의 claude -p로 넘기는 로컬 라우터 · Claude Code skill

Self-hosted interface for building and testing typed TypeSafe Jev API challenges.

Insurance claim triage console using TypeSafe AI's Jev model, with rule-based fallback, policy rules, tests and CI/CD

A jev intelligence layer for event stream to classify the app traffic into human vs ai_agent vs seo_tool vs scraper vs other_automation with bad_intent probability

Claude Code status line segment: Jev estimates turn progress (TypeSafe System One)

Prune-not-summarise context compaction for Claude Code, guided by Jev (TypeSafe System One)

Agent handoff: score companies against a clear problem using relevant evidence and compact shared-state Jev requests, with explicit cost accounting.

Jev-powered crypto signal classifier + paper trader using TypeSafe System One model

Simple Spring Boot demo for TypeSafe AI Jev

Jev (TypeSafe) decide qué modelo de Claude usar según la complejidad de la tarea

TypeSafe AI (Jev)를 학습합니다.

Distill Jev typed decisions into small portable task-specific classifiers (TypeScript CLI)

Learning TypeSafe AI's Jev evaluation model fast by building a /play request router for a DJ chatbot

No project description available.

Echo: an 87 KB copy of Jev for Snake. About 800× faster; with search, 5× the score.

Bulk Gmail cleanup pipeline: two-tier email classification (free heuristics + TypeSafe Jev System One model) with additive IMAP labeling. Classifies and labels — never deletes.

LLM-as-a-judge evaluation harness: a confidence-gated cascade (fast typed judge → Claude) that scores LLM outputs and gates CI on a regression diff. TypeScript, real recorded transcripts, real benchmark numbers.

Eval harness for TypeSafe Jev decision models: golden-dataset evaluation, calibration metrics, and schema-guarantee tests for typed decisions (Choice / Score / Noul)