jev-align (Sutro)

Active-learning CLI that uses human labels and GEPA to improve Jev decision definitions

Build calibrated AI Functions from human feedback using Jev and GEPA.

How it uses Jev

jev-align is an experimental CLI from Sutro for building AI Functions with decision models from TypeSafe, Perplexity, OpenAI, Liquid AI, and Cloudflare.

Getting started

Requires Python 3.11 or newer.

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README · the project website.

Keep exploring

Similar to jev-align (Sutro)

Browse all
typesafe-comment project preview
Applications
Small Python package that uses typesafe.ai to evaluate code comments on certain heuristics
jev-pr-labeler project preview
Applications
Semantic GitHub PR labels using Jev's typed decisions, with conceptual scope instead of line counts
llm2jev project preview
Applications
Any chat model, any engine (SGLang, vLLM, transformers) as a Jev-compatible probability decision service: one prefill, one label token
JevAny project preview
Benchmarks & research
Calibration-aware reinforcement learning for adaptive decision systems
JevHarness project preview
Agent tooling
LLM-authored task-specific Jev harnesses with optional full-trajectory reward reflection and GEPA evolution.
jev-align project preview
Benchmarks & research
Calibrated alignment verifier for LLM responses and agent plans — powered by Jev
foreman project preview
Agent tooling
Software Factory Foreman: an agent supervisor that uses Jev decisions to keep coding agents on task.
jev-router (prismhq) project preview
Open-source LLM router that uses TypeSafe's Jev to pick a model, on top of LiteLLM
jev-browser (KesavanKing) project preview
Browser & computer use
Local browser automation UI that uses TypeSafe Jev to choose bounded page actions and a text model only for field values.