Jev ecosystem projects — page 22

Pokemon Red on PyBoy: code owns the route and the arithmetic, Jev picks at branches in about 100 ms, calibration measured instead of assumed

Does a TypeSafe Jev rerank beat embedding search? Graded relevance eval (9,831 pairs, 164 zh/en queries) over the Agent Skills Hub catalog, with the judge-circularity bias measured.

Compile agent policy prose into deterministic verdict programs: narrow evidence questions for the model, the verdict computed in code. Install: npm i -g jev-compiler

Typed, calibrated decisions from a local model. No text generated.

daf-jev: composable Python toolkit for TypeSafe's Jev (System One) decision API — question builders, confidence gates, evaluator, calibration, CLI, MCP server, agent skill

Reproducible early-access evaluation of Jev on Korean understanding and medical text, with runtime and cost evidence

100 AI NPCs live in a tiny town. Jev chooses the next action; the world writes the story.

A small, fast prose linter: ruff-style rule codes for writing, backed by TypeSafe's Jev model

Screen a folder of CVs with the TypeSafe Jev decision model: typed judgments, an editable policy, free re-scoring.

Hybrid coding harness: System 2 writes, System 1 (Jev) runs reflexes.

Benchmarks and a playground for TypeSafe's Jev (System One) model: chess, and who-is-the-player-talking-to for speech-to-text game NPCs

A word-level language model whose output layer is Jev: n-gram drafter, Noul chunk verification, bits-per-token eval

Discriminative Monte Carlo Tree Search using System One and Harnesses

Orchestrate your own AI agents and build extensions for them — summon an agent by name, watch it think, ship its work. Runs on nostr, so no company owns your agents, your data, or your identity.

Make the model you already use work more like a frontier model with better planning, persistent context, skills, hooks, failure handling, and verification.. Orchestrated Multi-Specialist Agentic Lifecycle Harness

A verification layer for AI evaluations. Checks the instrument, not just the score: data, scorer, runs, numbers, claims, and itself.

A telegram bot which uses machine learning to detect spam messages

The decision layer for LLM apps

Tests, calibration audits and failure-mode studies of Jev (TypeSafe System One): jaggedness, consistency, injection, abstention.

Jev agent authorization for MCP tool calls: Kinde identity and permissions plus Jev's typed, calibrated decisions, checked server-side before every call runs

Curated list related to System One

Benchmark TypeSafe JEV against LLMs, fine-tuned BERT, Laya and zero-shot NLI on text classification: accuracy, calibration, latency, throughput, cost

The most complete gallery of what people build with Jev, TypeSafe's System One model: 3,400+ projects, demos, and write-ups by scenario, each with its original link, image, and description.