The boring part of an AI agent is the part that works

Description and use

Automating incident routing via email with an LLM and external directory reduced MTTR from 2 hours to 10 minutes.

Problem

Incoming incident emails are semi‑structured, noisy, and often misdirected, causing delays of up to an hour before a human can act.

Outcome

By normalising emails, constraining LLM output to a fixed enum, and delegating dynamic owner resolution to a directory service, the system achieved a ten‑fold MTTR reduction without retraining the model.

Workflow

• poll inbox
• normalise email body
• classify incident with LLM
• resolve owner via directory lookup
• create ticket
• notify owner
• observe outcome

Evidence

I built an agent to collapse that loop for a BSS platform, and it took mean time to resolution on routed incidents from around two hours to about ten minutes.

Source

The single highest‑leverage thing I did was aggressively strip the email before the model ever saw it.

Source

The model does one small job in the middle. Everything that makes the system trustworthy is plumbing.

Source