What challenges did Wonderful face before AI implementation?

Build reliable multilingual AI customer service agents that can handle real-time voice, text, and email interactions across many languages. Reduce contact-center costs while improving latency and containment. Support evolving enterprise workflows such as insurance claims and policy understanding.

What was the impact of the AI solution?

Customer service costs can be reduced by up to 70% according to the company. Early deployments helped contact centers handle thousands of simultaneous conversations. A banking client created a promotional agent in just four days, versus several weeks for 50 people. The company is scaling to more than 30 markets. AI agents can be deployed in days instead of months.

Source

GCPProductionEvidence: Medium65/100

Wonderful: multilingual AI voice/text agents for enterprise customer service using Gemini and Vertex AI

Use case typeCustomer service automationUpdated Jun 13, 2026

Wonderful builds enterprise AI agents that operate across multiple use cases, languages, and communication channels including voice, text, and email, adapting to local context. The company says it began its journey with Gemini and Vertex AI on Google Cloud to power a real-time multimodal voice pipeline, model orchestration, and retrieval/reasoning across enterprise knowledge and policy documents. Wonderful also uses Google Cloud infrastructure such as Compute Engine, Cloud Storage, and Google Kubernetes Engine (GKE), and applies a Zero Data Retention configuration so data is processed in-memory only.

Organization: Wonderful
Industry: Other
Location: France
Published: June 2026

Reported outcomes

−70%

costCost savings

Strategic outcomes

Customer experience & trustBuilt multilingual omnichannel customer service agentsNew product / capabilityEnabled real-time multimodal voice workflowsRisk & complianceAdopted in-memory data processing for retention controlMarket & geographic expansionScaling into more than 30 markets

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Cost: 70% decrease

Google Cloud Customer StoryJun 3, 2026Customer storyInferred claimMedium evidence strength

Customer service costs can be reduced by up to 70% according to the company.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Wonderful
Provider: GCP
Maturity: Production
Linked source: Google Cloud Customer Story

AI agents can be deployed in days instead of months

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 4

1Customer Service Automation
2Conversational AI
3Multilingual AI

Build reliable multilingual AI customer service agents that can handle real-time voice, text, and email interactions across many languages.
Reduce contact-center costs while improving latency and containment.
Support evolving enterprise workflows such as insurance claims and policy understanding.

Started with Gemini and Vertex AI to power real-time multimodal voice workflows and model orchestration.
Implemented a Zero Data Retention configuration so processed data stays in-memory only.
Scaled the platform on Compute Engine, Cloud Storage, and Google Kubernetes Engine (GKE).
Uses Gemini as an 'AI as a judge' to review conversations and improve prompts, and for retrieval and reasoning over multiple policy documents and knowledge bases.

Technologies

Gemini Vertex AI Compute Engine Cloud Storage Google Kubernetes Engine

Customer service costs can be reduced by up to 70% according to the company.
Early deployments helped contact centers handle thousands of simultaneous conversations.
A banking client created a promotional agent in just four days, versus several weeks for 50 people.
The company is scaling to more than 30 markets.
AI agents can be deployed in days instead of months.

Architecture

Wonderful's platform combines Gemini and Vertex AI for a real-time multimodal voice pipeline and model orchestration, then runs on Compute Engine, Cloud Storage, and Google Kubernetes Engine (GKE). The customer story also mentions a Zero Data Retention setup and Gemini-based conversation review and retrieval/reasoning across enterprise documents.

Sources & evidence1

Evidence: Medium65/100Evidence strength