Sai Pavan Kumar Kasaragadda

About

Backend/platform engineer with hands-on ownership of Kubernetes GitOps delivery (GitHub Actions + Argo CD) on AWS EKS, including progressive rollouts and reliable rollback across interdependent microservices. Built a Python/FastAPI ML-driven document-processing service (PostgreSQL + S3) to complement existing Spring Boot systems, and implemented Kafka streaming pipelines with Schema Registry plus Prometheus/Grafana observability. Also supported a hybrid cloud-to-on-prem migration for compliance/latency with phased rollout and incremental PostgreSQL migration.

Experience

Education

Key Strengths

• Designed and built Python FastAPI microservice for ML-driven document processing integrated with existing Java/Spring Boot services
• Production Kubernetes delivery on AWS EKS using Helm with autoscaling, resource limits, and tuned liveness/readiness probes
• Implemented GitOps CI/CD with GitHub Actions + Argo CD (ECR image build/push, automated sync, rollback)
• Resolved Argo CD multi-service dependency issues using sync waves and resource hooks
• Implemented progressive rollouts with Argo Rollouts and customized health/readiness checks to reduce false negatives
• Led hybrid cloud-to-on-prem migration planning and execution with phased rollout and incremental PostgreSQL data migration
• Built Kafka real-time streaming pipeline with focus on high throughput, consumer group/partition tuning, and schema governance via Schema Registry
• Set up observability for streaming and services using Prometheus/Grafana (lag/throughput monitoring and alerting)
• End-to-end ownership of enterprise web applications (requirements through deployment and support)
• Rapid iteration via small incremental releases and early stakeholder validation
• Architecting scalable React/TypeScript frontends with clear separation of concerns and reusable components
• Real-time UI performance optimization using Redux, memoization, and selective rendering
• Microservices at scale with RabbitMQ (ordering/reliability, DLQs, acknowledgments, idempotency)
• Proactive scaling and throughput improvements (consumer scaling, batching, monitoring/alerts, prefetch tuning)
• Built and drove adoption of an internal real-time analytics dashboard for ops/support, reducing manual effort and incident response time
• Strong prioritization and roadmap communication in ambiguous, shifting stakeholder environments