Serverless Sidecar Pattern for Offloading Auxiliary Logic

Had this idea and just getting it down on paper.

Concept

• Your core business logic lives in containerized application, running on Kubernetes. (your org or team is not fully bought in to cloud native and serverless :)
• Your write serverless functions (e.g. AWS Lambda) to handle auxiliary logic
• The serverless functions can be triggered via DynamoDB streams, events, etc.
• How far can or should this be taken?
• Serverless functions calling the core application’s REST APIs?