Which approach is correct for deploying Lambda functions in an AWS Elastic Kubernetes Service (EKS) cluster?

Deploying Lambda functions as part of an EKS cluster is a valid pattern when you want to combine container‑based workloads with serverless functions. In practice, you keep Lambda as a separate AWS service and invoke it from your pods using the AWS SDK or the Lambda API. EKS provides the orchestration for containers, while Lambda handles short‑lived, event‑driven tasks. You can also use the AWS Service Operator for Kubernetes to create Lambda resources directly from Kubernetes manifests, or run Lambda as a container image inside EKS if you need to share the same runtime environment. The key benefits are that you maintain the scalability and resilience of EKS for microservices, and you can offload bursty or compute‑intensive logic to Lambda without managing additional servers. This hybrid approach also simplifies cost management by allowing you to pay only for the compute you use in Lambda. Additionally, using Lambda with EKS enables event‑driven micro‑services that can scale independently of the container workloads.