jevFIELDNOTES / COMPARE
Decide where Jev belongs
Four different decisions: choosing an output type, comparing a GPT classifier, adding agent routing, or replacing rules. These are selection guides, not our own model benchmarks.
01 / 2 MIN READ
Jev or an LLM: start with the output you need
Use a fixed decision for routing and scoring; use generated text for replies and explanations.
Read guide →02 / 2 MIN READJev vs GPT for ticket classification
What the official demo shows, what it leaves open and how to compare your own classifier.
Read guide →03 / 2 MIN READWhere would Jev fit in a Claude agent?
Evaluate a routing step inside the agent rather than treating it as a replacement for the whole agent.
Read guide →04 / 2 MIN READJev, rules or a trained classifier?
Start with the simplest approach that handles your actual labels and mistakes.
Read guide →