jev-architect (BenjaminPolge)
Makes Claude Code and Codex ask whether a step needs a generative LLM at all — or whether it belongs on Jev, TypeSafe's System One model. Architecture arbitrage before the code is written.
How it uses Jev
When you ask a coding agent to build an AI feature, it reaches for the tool it knows — an LLM call — for every part of it, including the parts that are just a decision. Classifying a ticket, scoring a lead, routing a request, answering yes or no: those have closed outputs. They can run on Jev, TypeSafe's System One model, which returns typed values with probabilities instead of text, in the low hundreds of milliseco…
Highlights
- Runs before the code is written, at the architecture step.
- Splits a feature into atomic steps and assigns each one an engine: Jev , LLM , or
- Compares on eight axes — functional fit, cost, latency, volume, determinism, need for free
- Also works on existing projects : when you review or optimize a codebase, it flags LLM
- Stays quiet the rest of the time.
Getting started
Paste this into Claude Code or Codex:
- License
- Apache-2.0
- Last activity
- Sep 2026
- Added
- 2026-09-23
- Stars at snapshot
- 2
- Status
- Community listing
Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README.