What separates support AI agents that close tickets from ones that just deflect
Exact_Ostrich164 · reddit · 2026-09-23
The author outlines what separates support AI agents that actually resolve tickets from those that deflect until users demand a human.
Effective agents: call real tools (order tracking, account status, refunds) instead of only answering from a knowledge base; sit on top of existing helpdesks (Intercom Fin, Zendesk AI, and Aissist across Zendesk/Gorgias/Intercom/Freshdesk); enforce clear escalation rules with full context handoff; and start narrow (WISMO, resets) before expanding once resolution rates hold. The handoff to humans is where most setups break. The author asks practitioners for their real end-to-end resolution rates.
More from coding & agent
- Dev uses Opus 5.5 with Lean to formally verify Claude Agent SDK, shipping 16 bug-fix PRs — jimmykoppel · 2026-09-23
- Rogo CEO names memory compaction as the key unsolved problem for enterprise agents — rohanpaul_ai · 2026-09-23
- Graph Engineering: Building Reliable AI Agent Systems as Explicit Task Graphs — Pavan_Belagatti · 2026-09-23
- Have Your Coding Agent Attach Flame Graphs to Every PR It Opens — DanielLockyer · 2026-09-23
- Early hands-on: Sol 6 shows strength at goal-driven tasks, dev lets it run overnight — gregmushen · 2026-09-23
- Theorem says Lean-verified AI sandboxes are months away, at 1-30KB of proofs verified per hour — ctjlewis · 2026-09-23