How Engine Built AI Agent Eva with Agentforce to Resolve 50% of Customer Inquiries Autonomously

Engine is a travel platform built for small and medium businesses, handling hotel, flight, and car rental bookings for over 1 million travelers. As the company grew beyond its original hotel focus into a full-service operation — managing complex itineraries, group bookings, mid-trip changes, and cancellations — the volume and complexity of customer service interactions grew faster than the team could scale headcount to match.

The answer was Eva, an AI customer service agent built on Salesforce Agentforce and grounded in Data Cloud-unified customer profiles and booking data. Eva handles end-to-end travel requests across voice, web, and mobile channels — not just answering questions, but taking action: cancelling bookings, modifying reservations, sourcing alternatives, and routing genuinely complex cases to a human agent with full context already attached. The result is 50% of chat inquiries resolved autonomously by Agentforce and a 15% reduction in average handle time across the team.

In parallel, Engine deployed an internal Slackbot that gives every employee a personal AI assistant. The Slackbot takes skill-specific actions tied to each team's workflows, and 85% of employees use it — a number that represents adoption across every function in a company of over 1,000 people, not just technical teams.

The development infrastructure behind both products is as notable as the outcomes. Engine built its first AI agent in 12 days using the Agentforce 360 Platform's hybrid approach: low-code Agentforce Builder for configuring agent behavior, AI-powered Agentforce Vibes for code generation, and Apex for pro-code customization — all in one system with shared testing and deployment pipelines. This collapsed what would traditionally be a months-long agent development cycle into days, and gave Engine a repeatable process for building, testing, and releasing new agent capabilities as the platform scales.

Engine is now expanding Eva's scope — each new channel and booking type is a new agent capability deployable in the same hybrid framework — and using Tableau Next on top of Data Cloud to turn booking and engagement data into real-time insights for both product decisions and personalized customer communication.