A serverless workflow uses AWS Step Functions to invoke several Lambda functions. One Lambda intermittently times out during high load. The developer wants the workflow to automatically retry that function when a timeout error occurs. What change will accomplish this?

AWS Step Functions provides built-in error handling and retry mechanisms. By adding a Retry field to the state definition of the Lambda function that is timing out, you can specify which error types (like Lambda.Timeout or States.Timeout) should trigger a retry, along with parameters such as MaxAttempts, IntervalSeconds, and BackoffRate. This directly addresses the requirement for automatic retries on timeout errors. The other options are incorrect because: Adding a Timeout field to the state definition primarily controls how long the state is allowed to run before it times out, not how many times it retries. A Fail state terminates the workflow execution, which is the opposite of retrying. Using an SNS topic and a separate Lambda for retries introduces unnecessary complexity and doesn't leverage Step Functions' native retry capabilities.