I Built an AI Pipeline That Reads Support Emails and Drafts Replies (Here's What Actually Broke, and the Math on Whether It's Worth It)

Описание и применение

Пайплайн автоматически классифицирует и черновит ответы на поддерживающие письма, отправляя уведомления в Slack для проверки человеком. Стоимость API около $0.01 за письмо, в то время как ручная работа стоит $1.67.

Задача

Small business owners spend hours answering repetitive support emails, many of which are variations of a few common categories.

Результат

The pipeline cuts per‑email labor cost from ~$1.67 to ~$0.01 in API spend, while keeping human review. It handles routine high‑confidence cases automatically and routes ambiguous ones to human review, achieving a balanced automation ratio.

Рабочий процесс

• Pull unread Gmail messages via API
• Extract email body handling multipart/alternative and HTML fallback
• Classify with Claude using structured JSON schema (category, priority, intent, confidence, suggested_reply)
• Route to human based on category and confidence tier
• Generate draft reply in same API call
• Post Slack card with classification, draft, and assignee
• Mark email as read and record result in SQLite
• Human approves or edits draft in Gmail; no auto‑send

Доказательства

The system prompt is about 3,100 characters (~840 tokens), sent on every call along with the email body. At Claude Sonnet's current pricing ($3/MTok input, $15/MTok output), a typical classification-plus-draft call - roughly 1,050 input tokens and 200 output tokens - costs approximately $0.0064 per email.

Источник

Industry surveys on support ticket handling time put a straightforward inquiry at somewhere between 3 and 8 minutes of staff time... one manually-answered routine email costs roughly $1.67 in labor.

Источник