A company deployed its retail website globally. The site runs on multiple Amazon EC2 instances behind an Elastic Load Balancer in an Auto Scaling group across multiple Availability Zones. The company wants to serve different content versions based on the device used by customers. Which combination of actions should a solutions architect take to meet these requirements? (Choose two.)

To serve different content versions based on the device, a solutions architect should use Amazon CloudFront with Lambda@Edge. CloudFront can cache multiple versions of content, improving performance and reducing origin load. Lambda@Edge allows custom logic to be executed at CloudFront edge locations, enabling dynamic content modification. By inspecting the User-Agent header, a Lambda@Edge function can direct CloudFront to serve device-specific content from the cache or the origin. Incorrect options: Network Load Balancers (NLBs) operate at Layer 4 and do not support host-based or path-based routing, which are Layer 7 features. Therefore, configuring host headers or path-based routing on an NLB is not possible. AWS Global Accelerator improves performance by routing traffic over the AWS global network but does not provide content versioning or device-specific content delivery capabilities on its own. It would still require a mechanism like CloudFront and Lambda@Edge for that functionality.