Jev ecosystem projects — page 39

Ask Jev over columns of data: closed-set questions, cached and joined back.

Don't burn your model's context on a search flood — pipe it in and read above the cut line: one calibrated number per candidate, ~30 rows whether the pool is 40 or 4,000. It doesn't search; it decides what each hit is.

Coding-agent tools on TypeSafe's Jev, measured in public. review-router: caught 13 of 13 CVE fixes a filename rule sent to a quick review. jev-gate: a strict second lock, 0 of 12 dangerous test commands ran where Claude Code's auto mode let 8 run.

No project description available.

Email triage PoC: reads a mailbox over IMAP and classifies each message by what it is and what it's about, using TypeSafe System One (Jev) — 11 questions in a single call, decided in Python.

JEV-inspired local decisions for Apple Silicon, powered by MLX.

Model router for Claude Code using Jev

An OpenAI-compatible proxy that sits between your LLM and your users. It evaluates each sliding window of tokens while the response is still streaming and cuts the stream before a violating token can reach the screen.

Search code and text by meaning with natural-language grep, powered by Jev.

Rerank, filter, and compress JSON search results with TypeSafe AI's Jev.

Jev Resume Disqualifier: Sub-25ms automated resume knockout engine powered by TypeSafe Jev System One decision intelligence. Eliminates 80% of unqualified applicants with deterministic date math & EEOC-safe rejection notices.

Jev 1.13 reward-model evaluation across 8 benchmark tracks, with an interactive report and 54-row SOTA comparison

Local-first document evaluation workspaces powered by Jev

Jev-guided monkey tester: $0.35 and 5 minutes to find missing features, validation gaps and dead controls before QA

Per-turn skill selection for LangChain and deepagents agents: a fast judge picks the few skills a turn needs, so a catalog of hundreds stays out of the prompt.

Abandoned Qwen3.5-0.8B LoRA fine-tuning experiments for Jev-like typed judgments, with datasets, adapters, evaluations, and a full retrospective.

Binary approval gate for Codex and Claude Code — every intercepted tool call is reviewed by TypeSafe JEV and composed through a versioned local policy, with scoped authorization.

Jev-style typed decisions from Qwen3.5-2B logits — one forward pass, zero decoding, zero fine-tuning.

Community-maintained Java SDK for the TypeSafe AI System One (Jev) API. Not an official TypeSafe product.

Windows Codex Skill：通过 npx 或 Git 安装，安全配置 save-token-jev 的 PreCompact/SessionStart Hooks，并提供信任、原生压缩与旧内容隔离验证。

用 Jev 把飞书工作消息分成四类：紧急、待办、值得看、暂时略过。A minimal, local-first message classifier with extensible workspace adapters.

38 zero-shot information-extraction & classification systems across 23 families — extractors, classifiers, cross-encoder rerankers and typed-decision engines, local + hosted (OpenRouter) — demos, accuracy/latency/cost benchmarks, 9-language suite, web UI

A public wall where anyone asks a question in three to fifteen words and Jev, TypeSafe's judgment model, answers yes, no, or it depends in about 100 milliseconds. Every judged ask lands on the wall in realtime, with a running count toward one million, showing cost.

No project description available.