Worked on the kubernetes/kubernetes repository to deliver a memory-efficient transformer for the DeltaFIFO and Reflector components, targeting improved cache performance during periods of high object churn. The approach involved introducing new interfaces and methods to support object transformation within the caching pipeline, enabling objects to be transformed before storage and reducing memory bursts. By integrating the transformer from the provided store into internal stores within Reflector, the work enhanced the scalability and stability of the client cache layer. This feature was implemented using Go and Kubernetes, focusing on backend development to lower memory pressure and accelerate cache refreshes under load.