Carvana is reshaping its customer experience by pairing an AI agent named **Sebastian** with an AI-powered quality review platform called **CARE (Conversation Analysis Review Engine)**, all built on **Microsoft Azure**. Here’s how it works:

• Sebastian, the AI agent – Guides customers through the buying and selling journey, handling standard questions and tasks so customers get faster, more consistent support.
• CARE, the quality review platform – Analyzes millions of customer conversations every month to spot patterns, identify issues, and surface opportunities to improve both Sebastian and human Customer Advocates.
• 100% visibility into interactions – Instead of sampling a small subset of calls, Carvana now has full visibility into all customer interactions, which helps validate training efforts and keep service quality consistent.

This entire setup runs on the **Microsoft Azure AI ecosystem**, including:

• Azure AI Foundry to orchestrate AI capabilities.
• Azure AI Speech for fast, accurate transcription of calls.
• Azure Cosmos DB to store and scale conversation data securely.
• Azure Kubernetes Service (AKS) to reliably run and scale services.

By combining Sebastian, CARE, and Azure, Carvana can continuously refine its approach, deliver more personalized support, and keep improving the overall customer journey end to end.

Carvana’s AI-driven approach is delivering clear, measurable results across its operations. Key outcomes include:

• 45% reduction in inbound calls per sale over the past two years, as Sebastian has improved with ongoing refinement from CARE.
• More time for high-value work – Sebastian now handles more standard, repeatable scenarios, which frees Customer Advocates to focus on complex issues and more personalized support.
• Consistent service quality – With CARE providing 100% visibility into customer interactions, Carvana can confirm that training is working and that Advocates are aligned in how they serve customers.
• Faster product delivery – The engineering team uses GitHub Copilot to automate routine coding tasks, helping them ship features and improvements more quickly.

These improvements translate into a more efficient operation and a smoother experience for customers, while still keeping human experts focused where they add the most value.

Carvana chose **Microsoft Azure** as the foundation for Sebastian, CARE, and its broader AI strategy because it needed a secure, scalable, and flexible platform that could keep up with rapid innovation. Several factors drove that decision:

• Unified ecosystem – Most of Carvana’s systems already run on Azure Kubernetes Service (AKS), giving the team a single, integrated environment to build, deploy, and manage services.
• Speed and flexibility – With tools like Azure Cosmos DB, Carvana can adjust schemas, scale throughput and storage automatically, and move quickly on complex projects without knowing every requirement upfront.
• Security and compliance – Customer conversations can include sensitive data, so Carvana relies on Azure’s enterprise-grade security and compliance to ensure that information accessed by Sebastian and CARE is handled appropriately.
• AI-first capabilities – Azure AI Foundry, Azure AI Speech, and related services provide the workflows, models, and transcription capabilities needed to analyze millions of conversations in near real time.
• Future-ready architecture – Carvana is already testing Azure AI Foundry Agent Service to manage multiple AI agents with minimal code, positioning the company to keep reimagining how it supports customers.

In Carvana’s view, having its infrastructure and AI foundation on Microsoft is a competitive advantage: it helps the company run fast, adapt to market changes, and innovate with less complexity while maintaining strong security and reliability.