JevMade hello@JevMade.com
← Back to evaluation videos

JevMade field notes / Video guide

Jev rend tes appels d'IA INUTILES (30 000 tests)

Meydeey explains Jev’s bounded decisions and reports a 30,000-test comparison with a generative model. He reviews timing, costs, confidence, and failure cases, including repeatability and nested structures.

Original by Meydeey | Automatisation IAEvaluationIntermediate18 min 37 sec Published Source reviewed

Before you press play

What you’ll find in the video

  1. Jev returns typed decisions within supplied options rather than free-form text.
  2. Meydeey proposes routing, guardrail, and classification checks before sensitive agent operations or more expensive text generation.
  3. The creator reports failures on repeatability and nested-structure cases despite high schema compliance; valid structure does not establish a correct decision.
Worth knowing

Gemini-assisted video/transcript review. The figures come from the creator’s own test suite, which this review did not reproduce. Proposed guardrail uses and high confidence scores do not establish safe automated execution.

Jev rend tes appels d'IA INUTILES (30 000 tests)