Jev Customer Support

A Chinese-language e-commerce customer-service demo on FastAPI, LangGraph and React: adaptive retrieval routing with hybrid BM25-plus-vector search, and intent classification switchable between TypeSafe Jev's structured choice interface and DeepSeek.

A LangGraph agent for the Forward Deployed Engineer role that classifies incoming support tickets with a Jev decision node, auto-closes basic issues to Tier 1 and sends technical ones to an LLM summarizer. The README notes Jev is mocked by a standard LLM in this demo.

A Quarkus proof of concept that dispatches support tickets with Jev: one call answers urgency as a noul, the best-placed team as a choice and severity as a score, and low-confidence tickets go to review instead of being assigned automatically.

E-commerce customer-service quality sandbox: Jev judges intent, emotion, urgency and risk per message, an 8-rule decision table routes to queues with SLAs, and dangerous agent phrasing is blocked with rewrite suggestions before it is sent.

A general-purpose customer-service platform rebuilt from OpenAI's ChatKit sample: one Jev call returns intent, emotion, confidence, a recommended function and the right configured agent; a chat model drives MCP tools and built-in RAG answers from a knowledge base. Changing industry means editing YAML, not code.

Python intent classification for customer support: Jev picks from twelve intents and returns a confidence score plus a per-intent probability distribution, flagging anything below 0.8 confidence as needing clarification while routing stays configurable downstream.

A customer-service chatbot routed by Jev: one API call per turn asks every question in the decision tree up front and discards the unneeded answers — the docs' speculative fan-out — with a human-handoff Noul and a frustration Score riding along.