Product Support Assistant

A lightweight assistant that answers product FAQs, classifies support messages, and routes users to the right help — built to reduce ticket volume and improve response times.

Tech Overview

The app ingests a small product FAQ knowledge base, matches user intent, and returns answers with confidence scores. When uncertain, it escalates with helpful context for a human agent.

Code Sample

from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI()

class Query(BaseModel):
    message: str

@app.post("/ask")
def ask(q: Query):
    # 1) classify intent  2) retrieve FAQ  3) call LLM  4) return answer
    intent = classify(q.message)
    kb_hit = retrieve_faq(intent)
    answer = call_llm(prompt_from(kb_hit, q.message))
    return {"intent": intent, "answer": answer}