Routespring
AI agents for airline operations.
I’m building LLM and voice agents for exception-heavy work across crew travel, service recovery, payments, and hotel operations.
The operating problem
A difficult case rarely arrives as a clean request. It comes through a phone call, a changing schedule, a supplier response, or a source system that has not caught up yet.
Routespring handles booking, hotel coordination, crew rebooking, disruption recovery, payments, settlement, and reconciliation.
What makes an agent usable
Production agents need defined permissions, limited system access, evidence of completion, and a clear handoff to a person.
I am working on that verification layer. My TraceGuard research tests one way to check an agent’s work against the external state it was meant to change.
Earlier work
Before Routespring, I helped build Pine Labs, GlobalLogic, hCentive, and Gallop.ai. The company history has the details.