HariKube uses Kubernetes as query layer and control protocol for AI-era microservices

HariKube is a Cloud-Native Platform-as-a-Service middleware that repurposes Kubernetes Custom Resource Definitions as a data query and application interface layer for managing distributed workloads. The platform leverages storage-side filtering against distributed databases such as TiDB and CockroachDB, turning the Kubernetes API schema into a unified representation layer for most service workloads. Its horizontally scalable control plane handles state synchronization, resource-based data sharding, and real-time event routing at large scale without dataset size constraints. A forthcoming experimental feature will allow Kafka producers and consumers to register directly into the middleware, enabling application state updates through data streaming and WASI plugin-controlled event flows. HariKube is positioning itself as a single source of truth for application state, targeting developers building microservices, AI agent platforms, and secure sandbox environments on Kubernetes.