What is Pinterest (NYSE:PINS) doing with AWS?

Pinterest announced a $4 billion expansion of its collaboration with Amazon Web Services (AWS) through 2031, marking the largest infrastructure investment in the company’s history. The goal is to modernize its platform architecture and accelerate AI capabilities.

Why is Pinterest investing heavily in AWS cloud services?

The investment supports Pinterest’s AI roadmap, enabling faster large language model (LLM) training, vision-language model execution, and improved performance of features like its Taste Graph and visual discovery tools.

What AWS technologies will Pinterest use under this deal?

Pinterest will leverage custom silicon processors Graviton and Trainium and migrate workloads to a Kubernetes-based environment using Amazon Elastic Kubernetes Service (EKS) for flexibility and scalability.

How does this affect Pinterest’s AI capabilities?

The upgrade will allow Pinterest to train and execute advanced AI models more efficiently, improving recommendations, search, and shopping experiences for its 600+ million monthly active users.

How long is the Pinterest-AWS agreement for?

The expanded agreement runs through 2031, giving Pinterest a multi-year infrastructure commitment to scale AI and platform services.

Tech

Image for illustrative purposes only. Not a real photo.

Pinterest inks $4B cloud services deal with AWS

Written by Brie CarterPublished Jun. 4 2026Last Updated Jun. 4 2026

Pinterest (NYSE:PINS) today announced a major expansion of its long-standing collaboration with Amazon Web Services (AWS), entering into a planned $4 billion commitment for cloud services through 2031.

This agreement marks the largest infrastructure investment in Pinterest’s history and is designed to modernize the platform’s underlying architecture while accelerating its roadmap for artificial intelligence.

The deal builds on a partnership that began in 2010, positioning AWS as Pinterest’s Preferred Cloud Services Provider.

Under the expanded agreement, Pinterest will leverage a variety of AWS technologies—including custom silicon processors such as Graviton and Trainium—to support its AI operations.

Pinterest intends to utilize AWS Trainium to train and execute the large language models (LLMs) and vision-language models that power its "Taste Graph" and visual discovery features.

The company is currently migrating from traditional EC2 environments to a more flexible, Kubernetes-based architecture using Amazon Elastic Kubernetes Service.

These advancements are expected to improve the speed and responsiveness of search and shopping experiences for the platform's more than 600 million monthly active users.