Pariveda is offering a 6-12 week use-case implementation to demonstrate how quickly predictive analytics use cases can be enabled. By leveraging ready to deploy CDK code to accelerate CI/CD, error handling and alerting, ML sandbox environments, data ingestion, data validation, and feature engineering, customers can focus on actually building predictive models.

Prior to delivery, an ideation workshop will help identify and select a valuable use-case, if one has not already been identified, for implementation of a predictive model. The selected use-case will be implemented and delivered to end-users.

Deliverables the customer will receive include:

• Prioritized use-cases and dependencies where a predictive analytics solution will provide value
• Maturity mapping of use-case expanded capabilities and value assessment of future phases.
• Networking requirements to complete use cases
• One change data capture data source (CDC) using the Database Migration Service (DMS)
• Inclusion of one reference data source into data product
• Batch based data pipeline delivered to data product
• Historical data stored in raw format in data lake
• Deliverable to include business use case, production data product, and solution financial modeling
• Data engineering capability that can be applied through a configurable solution to enable other use cases

AWS products and services used

• Amazon S3
• AWS Glue
• Amazon SageMaker
• AWS Step Functions
• AWS Lambda
• Amazon DynamoDB
• Amazon Athena
• AWS Code Commit
• AWS Code Pipeline
• AWS Code Build
• AWS Code Deploy

Target Customers for this Offering Target customers for this offering are companies that are looking to accelerate their use of AI/ML. The majority of tasks to put AI/ML into production have little to do with developing the models themselves.