Automated Student Data Pipeline with Kubernetes Deployment

Design a robust Kubernetes deployment strategy for a scalable student data processing microservice using Flask and pandas. Create a Helm chart that supports dynamic scaling based on data ingestion volume, implements multi-stage Docker builds with Alpine Linux, and includes comprehensive health checks and readiness probes. The solution must handle concurrent student record transformations from multiple institutional data sources, with built-in error handling and logging using structured JSON formats.