Fraud Detection · Marketing Personalization
iFood needed AI infrastructure that could personalize the ordering experience for tens of millions of monthly orders, improve fraud prevention, and standardize model implementation work for internal developers at marketplace scale.
Reported impact
iFood uses AI at the scale of more than 80 million monthly orders and more than 330,000 partner establishments, routing each purchase through more than 100 AI models to personalize recommendations, improve fraud prevention, and speed internal AI development.
iFood uses AWS generative AI and ML services to support two high-impact areas of its marketplace: customer experience personalization and internal developer productivity. Amazon Bedrock and Amazon SageMaker give the company standard building blocks for model implementation, experimentation, and deployment across recommendation, fraud-prevention, and delivery-experience workflows. When users open the iFood app and place orders, they are routed through a large portfolio of AI models that tailor recommendations and operational decisions at scale. The same AWS services also help the engineering organization standardize how new AI features are introduced so teams can move faster without rebuilding core infrastructure. This combination lets iFood apply AI throughout the full order journey, from discovery to delivery, while supporting the operational complexity of a national marketplace.
User interactions and order journeys in the iFood app pass through AI workflows built on Amazon Bedrock and Amazon SageMaker, which power personalization, fraud prevention, and developer-facing model implementations across the marketplace.
Only with AI can I create a personalized experience for each user, from recommendation to fraud prevention.
more than 330,000 partner establishments
more than 80 million orders per month
uses services such as Amazon Bedrock and Amazon SageMaker to ... optimize the customer experience ... and in fraud prevention processes.
when a user enters the app and makes a purchase through iFood, they go through more than 100 AI models.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
NVIDIA Sample Aug 14, 2026 Shopee Shopee needed large language models that could handle multilingual ecommerce tasks across Southeast Asia, Taiwan, and Brazil, but general-purpose frontier models degraded sharply on regional languages, shopping… 113x growth from 3 billion to 340 billion monthly tokens Compass v3.5 · NVIDIA A100 GPUs · NVIDIA H100 GPUs 
Databricks Sample Oct 14, 2024 Simpl Previously hampered by an outdated tech stack and siloed data, which caused frequent failures and lengthy ML model training, Simpl faced substantial scalability and operational challenges. 35% improvement in the analytics team’s turnaround time for reporting tasks Databricks SQL · Delta Lake · Lakeflow Jobs 
AWS Sample Feb 6, 2025 Mercado Libre High-quality visuals significantly shape a potential customer’s experience, yet creating compelling display ads remains a challenge for many vendors. Considering that 80% of Mercado Libre’s sellers are SMBs, they were… 7 countries Amazon Bedrock · Amazon DynamoDB · Amazon S3 