What challenges did iFood face before AI implementation?

Improve customer experience across a large-scale food delivery platform. Support AI-driven innovation across millions of monthly orders. Improve delivery service-level agreements and strengthen security and fraud prevention.

What was the impact of the AI solution?

The case study says iFood reduced costs by 40% using Kubernetes on AWS. The solution supports personalized AI experiences and scalable operations across Latin America. The platform processes up to 60 million requests per minute.

Source

Evidence: Medium50/100

iFood uses Amazon Bedrock to build a virtual waiter and applies Amazon SageMaker for delivery improvements

Use case typeCustomer service automationUpdated Jun 13, 2026

iFood is a Latin American online food delivery platform serving more than 300,000 restaurants and millions of orders each month. The company established an internal AI team and uses AWS services to standardize implementations, improve customer experience, and support AI-driven innovation across ordering, delivery, and fraud prevention workflows. iFood built a virtual waiter named Garçon using Amazon Bedrock, including fine-tuning mentioned by the company, and uses Amazon SageMaker to improve delivery service-level agreements, bolster security, and enhance customer and restaurant experiences. The company also trains internal teams through AWS Training and Certification to build machine learning, architecture, and security skills.

Organization: iFood
Industry: Consumer & Food
Location: Brazil
Published: May 2026

Reported outcomes

−40%

cost reductionCost savings

60 million requests per minuterequest throughput

Strategic outcomes

New product / capabilityBuilt a virtual waiter experienceNew product / capabilityImproved delivery and fraud workflowsCustomer experience & trustEnhanced customer and restaurant experiencesScale & capacityEnabled scalable operations across Latin America

Catalog median for cost savings deployments: −40% across 177 reported metrics. Compare benchmarks →

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

Normalized claim

Cost reduction: 40% decrease

AWS Customer StoriesMay 27, 2026Customer storyExplicit claimMedium evidence strength

reduced its costs by 40% using Kubernetes

Normalized claim

Request throughput: 60 million requests per minute

AWS Customer StoriesMay 27, 2026Customer storyExplicit claimMedium evidence strength

process up to 60 million requests per minute

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

Customer: iFood
Provider: AWS
Maturity: Unknown
Linked source: AWS Customer Stories

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 5

1Customer Service
2Operational Efficiency
3Fraud Detection

Improve customer experience across a large-scale food delivery platform.
Support AI-driven innovation across millions of monthly orders.
Improve delivery service-level agreements and strengthen security and fraud prevention.

Established the iFood Artificial Intelligence team to accelerate innovation.
Built the Garçon virtual waiter chat experience with Amazon Bedrock.
Used Amazon SageMaker to improve delivery SLAs, bolster security, and enhance experiences for customers and restaurants.
Upkilled internal teams through AWS Training and Certification.
Used Amazon EKS to process high request volumes in the platform environment.

Technologies

Amazon Bedrock Amazon SageMaker Amazon EKS AWS Training and Certification

The case study says iFood reduced costs by 40% using Kubernetes on AWS.
The solution supports personalized AI experiences and scalable operations across Latin America.
The platform processes up to 60 million requests per minute.

Sources & evidence1

Evidence: Medium50/100Evidence strength