Guides & comparisons
Practical guides on designing Jev questions, routing tickets, and choosing Jev vs an LLM for classification.
Jev Message Intent Detection: A Complete Playbook for Chat Classification
How to classify chat message intent with Jev — genuine question vs passive-aggressive vs casual chat — including state design, option wording, threshold tuning, and production fallbacks. Works as a drop-in alternative to LLM prompt parsing for high-volume message intent detection.
How to Route Support Tickets with Jev (Instead of Brittle Keyword Rules)
Replace keyword-based ticket routing with probabilistic classification. A practical guide to queue design, confidence thresholds, human fallback, and monitoring — for support teams that need accurate ticket triage at scale without LLM latency.
Jev vs GPT for Classification: When to Use Which (Cost, Latency, Output Contract)
A practical comparison of Jev and GPT-class LLMs for text classification: latency budgets, cost per 1k/1M decisions, typed outputs vs prose parsing, and a decision checklist. For engineers choosing a classification API for routing, triage, or tagging.
Jev API Tutorial: Getting Started with OpenRouter (Keys, Requests, Errors, Retries)
Step-by-step Jev API tutorial — get an OpenRouter key, send state + questions, parse typed answers, handle 429/400/401, and implement retries safely. Includes curl and TypeScript examples for production clients.
How to Design Effective Questions for Jev (Options, Granularity, Thresholds)
Best practices for writing Jev questions: option wording, choice vs score vs noul, why 'Other' is usually a bad idea, granularity rules, and threshold design. Practical patterns for classification prompts that actually map to code branches.