SpadinaBus Reciprokal

The purpose of this project is to construct a cloud-oriented offer serving engine that is highly available, scalable, configurable and maintainable. An endpoint must be exposed for applications and websites to integrate with to retrieve offers for logged-in customers.

• Offers are assigned to customers using other marketing software that produces an offer-to-customer allowlist CSV file. This allowlist maps an offer_id to a customer_id along with a score used to prioritize the best offers for each customer. This CSV is loaded into a highly available datastore (AWS DynamoDB) for querying.
• Offers are selected based on score and various configurations defining limits on viewing of offers by a customer. When a best offer is selected, a view is logged for it to the offer datastore.
• The offer data is then returned to the client who will present the best offer based on the offer_id.

Click h full project documentation, see: http://www.spadinabus.com/wp-content/uploads/2021/09/SpadinaBus-Offers-Engine-Technical-Documentation-v1.0.pdf

Contents of Repository

• 2 python lambda functions
  • get_offer_lambda.py endpoint handler to retrieve offers for a logged-in customer.
  • upload_offer_lambda.py s3 bucket update handler to upload new offer allowlist CSVs to Dynamo.
• Config files used in deploying this app (IAM policies, sample offers allowlist CSV, sample get_offer_lambda config)
• Notes such as helpful CLI commands and examples responses/structures returned by Dynamo and Lambda
• UML source diagrams (in .uxf format for UMLet)
• Code for an example demo website that integrates with Reciprokal and AWS Cognito