Live radar · scored by Jev
What people are building with Jev.
119 signals · last crawl
Jev Bookkeeping Automation
Jev is an independent index that automates bookkeeping by performing actions on a desktop accounting application based on a decision log and a dynamically rewritten playbook.
Open Source Jev Clone
This project is an open-source clone of the Jev API, designed to outperform the original in its core function by offering offline execution, zero cost, and open weights, while achieving competitive accuracy and significantly lower latency.
Shipped sqlite-jev: loadable sqlite extension + Python wrapper for asking @typesafeai #Jev typed
Noul, Choice, and Score via scalar functions and batched virtual-table queries, with confidence attached. https://github.com/mgaitan/sqlite-jev https://x.com/tin_nqn_/status/2101030999945580598/photo/1
AI-Powered Puzzle Game Bot
A bot plays a puzzle game by querying an AI model for tap probabilities before each move, utilizing live API calls for decision-making.
dre: A Keyboard-First Diagram Editor Driven by AI
dre is a diagram editor designed for keyboard control, which also enables AI to drive its operations by selecting from predefined choices rather than generating raw text.
Jev Performance on Ruka Workload
Jev processed 500 products with 801 possible master supplies in 8 seconds, achieving approximately 62 products/sec, a significant improvement over the current LLM pipeline's 2 products/sec.
Jev and GPT-6 Astra on Agilex PiPER Arm
This project integrates Jev and GPT-6 Astra on an Agilex PiPER robotic arm, with a process involving model skill selection, governor checking, and arm movement.
1/7 I'm building Claude Crossing, a cozy island life-sim, with a team of Claude Opus 5.2 <nex
1/7 I'm building Claude Crossing, a cozy island life-sim, with a team of Claude Opus 5.2 <next> agents. For the small judgment calls, I used @typesafeai 's Jev model: 2,162 calls, 54,851 questions, $0.39 in total.
TypeSafe AI Jev System One Model
TypeSafe AI's Jev is a System One Model that takes context/state as input and returns fast, typed probabilistic decisions for software to act on, rather than generating strings like traditional LLMs.
Jev Expense Categorization
This project uses Jev to automatically categorize expenses within an open-source application.
Agent Skill Categorization with Jev
This project utilized Jev to categorize a large number of agent skills into predefined categories.
Effectuai SDK Decision Model Support
The Effectuai SDK now supports Typesafeai's decision model JEV, with capabilities inspired by Tim Smart's implementation for the EffectTS AI package.
End of radar