New pattern: aws-apigateway-sagemakerendpoint

[dscpinheiro]

This new pattern is similar to the existing aws-apigateway-sqs pattern, but the service integration is with a SageMaker endpoint. The experience will be similar to the one described on this blog post: https://aws.amazon.com/blogs/machine-learning/creating-a-machine-learning-powered-rest-api-with-amazon-api-gateway-mapping-templates-and-amazon-sagemaker/

The main difference with this pattern is that only a GET method is made available (which will call the invoke-endpoint API). Also, since each ML use case is unique, the construct will require certain properties (such as resource path and request mapping template) to be provided.

Use Case

API Gateway provides a layer of abstraction from the endpoint. This layer of abstraction enables custom authentication approaches and control quotas for specific consumers.

Proposed Solution

Name	Type
apiGatewayProps?	api.RestApiProps
resourcePath	string
endpointName	string
requestMappingTemplate	string
responseMappingTemplate?	string

Example usage:

new ApiGatewayToSagemakerEndpoint(this, 'ApiGwSagemaker', {
    apiGatewayProps: {},
    resourcePath: '/predicted-ratings/{user_id}',
    endpointName: 'some-name',
    requestMappingTemplate: 'some-template',
    responseMappingTemplate: 'some-template'
});

• 👋 I may be able to implement this feature request
• ⚠️ This feature might incur a breaking change

---

This is a 🚀 Feature Request

[hnishar]

Published the pattern in the latest release v1.68.0