Airline Customer Support Ticketing System with AI Copilot

Overview

An enterprise-grade backend system designed to manage post-booking airline customer support operations. The platform handles passenger issues such as delays, cancellations, baggage problems, and refund requests, and is enhanced with an AI Copilot to support both passengers and customer support agents.

The Problem

Airline customer support teams handle high volumes of requests that require: Accurate flight and reservation context (PNR-based) • Proper prioritization and SLA tracking • Consistent and policy-compliant responses • Efficient agent workflows Traditional ticketing systems are largely manual and context-poor, resulting in slow response times and inconsistent handling.

The Solution

Clean Architecture–based backend with clear separation of concerns • Domain-driven ticket lifecycle with strict state transitions and audit logging • PNR-aware flight and reservation context for accurate categorization and prioritization • AI Copilot integration to assist passengers and agents with context-aware guidance • Background processing for SLA monitoring and escalation

Key Features

• ▸Role-based access (Passenger, Support Agent, Admin)
• ▸Controlled ticket lifecycle with audit trail
• ▸SLA monitoring and priority escalation
• ▸AI-assisted ticket creation and response drafting
• ▸Knowledge base powered by RAG for policy consistency

Architecture

Backend: ASP.NET Core Web API (.NET 8/9) • Architecture: Clean Architecture + Domain-Driven Design • Data: SQL Server, Entity Framework Core • AI Layer: LLM-based Copilot with RAG-enabled knowledge base • Background Services: SLA monitoring and notifications • Security: JWT authentication, role & policy-based authorization

My Contribution

Designed and implemented the backend using Clean Architecture principles • Modeled the ticketing domain, lifecycle rules, and role-based permissions • Implemented PNR-based flight and reservation context handling • Integrated AI Copilot for passenger self-service and agent decision support • Built background services for SLA monitoring and escalation • Prepared seeded data and API documentation for realistic demo scenarios