Live radar · scored by Jev

Jev use cases, projects and experiments

Real builds with TypeSafe AI's Jev — scored, not liked.

New here?Jev isn't chat →

1,857builds on the radar

last crawl 4d ago

Patterns rising

Shared ideas across independent builders

See all

Crash datasets that carry an investigator narrative hold information the coded fields omit. Coding those narratives at scale has been blocked by three obstacles. Frontier large language models are costly at that scale, their generated text cannot be verified, and no rule says how much output a human must check. This paper formulates narrative coding as gated, typed decisions answered by Jev, a System One model that returns probabilities over analyst-defined options and generates no text. A screen covered 499,500 Texas narratives and 195,857 were coded with a 27-question schema. Cost is governed by schema size rather than narrative length. The probabilities are audited against coded fields and against 2,416 blinded human judgments drawn under a stated sampling design. Two frontier large language models are benchmarked on the same records. Against human labels the typed model attains an F1 of 0.908. One frontier model gains 0.059 and the other is indistinguishable from it. Calibration va

Police Crash Narrative Conversion to Probabilistic Variables

This project converts police crash narratives into probabilistic crash variables using a System One Model, referred to as Jev.

Natural Language Processing Papers@HEI

38 views on XJev score 78

Jev System Prompt

This project provides a system prompt designed to architect a control system around a large language model (LLM), separating generation, decision-making, and deterministic code execution to optimize LLM usage.

God of Prompt@godofprompt

3K views on XJev score 75

Agent Decision Layer

This project implements a decision layer that sits before expensive agent actions, determining whether to act, escalate, or involve a human based on confidence scores.

yankis@yankis0x

452 views on XJev score 79

Jev Traps

Jev Traps is an open-source content firewall for AI agents that classifies intent to mitigate risks before untrusted content reaches the agent.

Sal@cyberesian

351 views on XJev score 73

Vectorless RAG with Jev

This project introduces a novel approach to Retrieval Augmented Generation (RAG) that replaces traditional embedding methods with a structured, table-of-contents-based navigation system, significantly improving speed and reducing costs by offloading non-generative tasks to Jev.

Oludele Halleluyah@hallelx2

155 views on XJev score 82

Jev vs. LLMs in Snake Game

This project is a small experiment comparing the performance of a system named Jev against six large language models (LLMs) playing the game Snake, measuring points scored and decisions made within a 60-second timeframe.

Angel Galvis Caballero@angelgalvisc

154 views on XJev score 72

Jev Decision Model

Jev is a typed decision model that returns a probability for a given state (question, answer, evidence) instead of generating prose or structured data, making it faster and cheaper than traditional AI judges.

Jose Luis Latorre | #MVP | #K6 | @DotNetZurich@joslat

40 views on XJev score 79

Jev Judgment Model Evaluation

Jev is a judgment model from TypeSafe AI that provides probabilities and confidence scores for typed questions based on a given state, designed to function as a non-acting governance layer.

Marquise@marquisehurtt

403 views on XJev score 80

JEV Fuse

JEV Fuse is a governed execution layer that translates raw decisions from probabilistic models into deterministic actions like ALLOW, ASK, or DENY, acting as a proxy between decision models and agent tooling.

0xShikhar⚡️@0xShikhar

41 views on XJev score 71

Why Didn't AI Cite Me?

This tool analyzes AI-generated answers and their citations to determine why a specific webpage was or was not used as a source, comparing its content against the AI's output.

Hrishabh Jain@hrishabh_hj

54 views on XJev score 69

Jev Decision Engine

Jev is a lightweight decision engine designed to score, classify, route, and decide on options based on provided criteria, offering structured scores, consistent signals, and fast, cost-effective processing.

Ujjal Bhattacharjee@ujjalcal

93 views on XJev score 69

Jev Model Asymmetry Analysis

This project analyzes the Jev model's behavior, revealing that natural language wording induces a significant conviction asymmetry between buy and sell sides, while the model itself is symmetric when presented with symmetric prompts.

Dan Holzrichter@dholzric

83 views on XJev score 79

Jev Decision Model Evaluation

Jev, a decision model from TypeSafe, was tested on standard AI problems, answering structured questions with probabilities and performing comparably to or better than other models on single-step tasks, but struggled with sequential reasoning without state information.

Nikhil Deshpande@nikhdesh

113 views on XJev score 76

TypeSafe Computer Use

This project enables a Mac to achieve plain-English goals by combining OCR and the Accessibility tree to drive actions, avoiding large model loops for screenshots.

Yash Chaudhary@YashChaudhary

4 views on XJev score 78

Jev: A Stable and Cost-Effective Judge for Agent Evaluation

This project explores Jev, a TypeSafe model, as an alternative to traditional LLM-based judges for evaluating AI agents, demonstrating significantly improved stability, reduced cost, and lower latency in scoring agent performance.

dante@danteisshipping

90 views on XJev score 71

Jev and Laya Pre-Request Gates

This project evaluates Jev and Laya, AI models designed for pre-request classification, comparing their performance against existing systems in speed and accuracy for tasks like credential detection and prompt injection.

Ran Aroussi@aroussi

300 views on XJev score 74

Jev Decision Model Playbook

This project provides a 10-page playbook to understand and implement Jev, an AI model designed to choose the next move for agents, acting as a decision-making layer rather than a generative one.

Hamza Khalid@humzaakhalid

433 views on XJev score 70

Jevmaxxing Agent

This project uses Jev to selectively classify search results, gating out a significant portion before they enter context, demonstrated by a real research session where 277 items were judged and 65% were gated out at a cost of $0.007.

⚡️ nedzen.eth@nedzen

152 views on XJev score 68

JEV for Interactive Browser UI

This project explores the effectiveness of JEV (likely a browser automation tool) for handling interactive browser elements like buttons, forms, and dynamic UI, distinguishing its utility from static web fact extraction.

BlackwellBoy@Blackwellboy

3.2K views on XJev score 68

Jev: Independent Index for API Pricing

Jev is an independent index that determines the value of an API answer by calculating expected value across a distribution of probabilities, rather than rounding to a fixed tier, enabling fairer and simpler pricing.

Tollstile Lab@tollstile

38 views on XJev score 76

Jev-Mem: Agentic Memory Stack

Jev-Mem is an agentic memory stack where a System-One controller manages typing, relation building, query routing, budget, scoring, and adaptive stopping, while System Two solely synthesizes information.

Sai Teja@saiitoshii

8 views on XJev score 73

TypeSafe Jev Skill Router

This project built a skill router where TypeSafe Jev selects the appropriate installed skill for each request, keeping the catalog out of context.

艾特玖@moziho592419

8 views on XJev score 75

Kev-0.6B ANE Performance Measurement

This project measures the performance of Kev-0.6B, a Jev-class decision model, running on the Apple Neural Engine, detailing its latency and power consumption.

Nick Lo@NickLo641579

71 views on XJev score 71

Computer-Use Agent Cost Optimization

This project optimizes the cost of computer-use agents by shifting from complex planning to closed-set choice, significantly reducing per-decision costs.

creedants@creedants

96 views on XJev score 71

What is Jev?

Jev is TypeSafe AI's System One model for fast, typed probabilistic decisions. Instead of generating prose, Jev can answer structured Choice, Score and Noul questions that software can act on directly. Jev Radar tracks real projects, experiments and production use cases being built with Jev.