HyunjunJeon/jev-judgment

HyunjunJeon/jev-judgment on GitHub

Stars: 3
Forks: 1
Language: Python
License: MIT
Last push: 17 Sep 2026

Read from GitHub on 24 Sep 2026. Counts change daily.

How it uses Jev

ENV_MODEL = "TYPESAFE_DEFAULT_MODEL"
ENV_BASE_URL = "TYPESAFE_BASE_URL"
DEFAULT_MODEL = "jev-latest"
DEFAULT_BASE_URL = "https://api.typesafe.ai"
DEFAULT_TIMEOUT_S = 8.0
DEFAULT_STRING_CHARS = 2000
DEFAULT_STATE_CHARS = 8000
QUESTION_TYPES = ("noul", "choice", "score")
# HTTP statuses with a specific `error.kind`; everything else is http_error.
_HTTP_KINDS = {401: "unauthorized", 422: "rejected", 429: "rate_limited"}

# ---------------------------------------------------------------------------
# Redaction. Ordered; first match wins per span. Deliberately conservative:
# a false positive costs Jev some context, a false negative ships a secret.
# ---------------------------------------------------------------------------

View in skills/jev-judgment/scripts/jev.py

Open on GitHub

PrimitivesChoice Score Noul

Use casesConfidence-gated actions

jev-trader

Makes one AI trade decision every Monad block on the Kuru MON-USDC market using Jev. The public deployment runs in dry-run mode with a mock model.

jarrodwattsOpen

2.2k starsLoop, mock model p50 100 msDecision window ~300 ms/block

ProjectConfidence-gated actions+1

jkudish/jev-mcp

MCP server that gives Claude Code, Codex, Amp, OpenCode and other agents ten Jev judgment tools, each returning typed probabilities they can branch on. The author calls it early software.

jkudishOpen

320 starsPer call 150-500 ms

ToolAgent routing and skill selection+2

AnyJev: Jev-style decisions from any open LLM

Python library from Nokia researchers that reads typed Choice, Noul and Score answers with probabilities from an open LLM's logits or hidden states, without fine-tuning. Its zero-label level reduces option-order bias, and a head fit on 100 to 300 labels can run behind vLLM.

Avi Chawla (@_avichawla)696 starsOpen

AlternativeConfidence-gated actions

Foreman

Places Jev as a fast supervisor above slower Codex coding agents. Codex workers do the engineering while Foreman independently assesses whether the work is complete, requirements are met, tests suffice, or human input is needed.

thruwire538 starsOpen

ToolConfidence-gated actions+2

kyotofin/tax-doc-classifier

A TypeScript classifier that sends each PDF page's text to Jev as a choice over 261 IRS forms and 7 page kinds, with a second call only for five corporate forms and their schedules. Its eval reports 0 wrong pages on 314 filled TaxCalcBench forms at about $0.001 per page, 34x cheaper and 6x faster than the Sonnet pipeline it replaced.

Nakshatra Saxena (@nedwize)Open

419 starsCost per page $0.001vs prior setup 34x cheaper

ProjectHierarchical classification+1

decider: one-pass typed decisions with calibrated probabilities (Qwen3.5-2B)

Open reproduction of the System One model class, fine-tuned from Qwen3.5-2B-Base, returning a probability distribution per typed question in a single forward pass. Serves TypeSafe's POST /v1/systemone wire format so TypeSafe SDKs work unchanged by repointing TYPESAFE_BASE_URL.

Mapika348 starsOpen

AlternativeIntent and model routing+3

Menu