JEV-Routed Assess-Decide-Do Workflow

Dragos Roua paired his “Assess Decide Do” cognitive framework with a JEV implementation. Assess and Do are LLMs that output text, while Decide uses JEV to evaluate a state dynamically updated by the LLMs and route to Assess, Do, or Done.

View on X

Dragos Roua@dragosroua𝕏

I paired my cognitive framework "Assess Decide Do" with a JEV implementation. In the movie below, "Assess" and "Do" are LLMs, they output text. "Decide" is JEV, it evaluates the state (which is dynamically updated by LLMS) and route to: - Assess, if we need more clarity, so specs are updated - Do, if we have clarity and just need to write code - Done, if the code manages the specs and the tests ar

Sep 20, 2026X postsView on X

Also filed under Agents & browsers

Evaluates Tool-Using Agents with Jev in CI
The project evaluates a real tool-using AI agent using code checks, a calibrated Jev judge, and an LLM judge. It is wired into CI.
Watchdog for AI Coding Agents
Jev Watch is a watchdog for AI coding agents that stops Codex or OpenCode when they loop, stall, or drift off task, then resumes the same session with a correction.
Fraud Investigation Agent
The author is building a fraud investigation agent using Jev and TigerGraph. Its flow goes from an alert through graph evidence, a Bayesian update, a pattern, Jev’s extra lookup, policy, a verdict, and human approval.
Coding-Agent Runtime Using Jev Decisions
Distill is a coding-agent runtime that uses Jev decisions for model routing, context retention, and tool selection.