Harshvardhan J.

Senior DevOps and Platform Engineer — AWS and Azure, Kubernetes, Terraform. Certified Kubernetes Administrator (CKA) and HashiCorp Certified Terraform Associate. Top Rated, 100% Job Success, 660+ hours delivered on Upwork. I build the infrastructure production systems run on, and I keep it running. EKS and AKS, reusable Terraform modules across dev/test/prod, GitOps with Argo CD, CI/CD pipelines with scanning built in, and observability that answers the question you actually have at 3am. Most recently I moved a healthcare platform's full production workload from AWS to Azure in a single planned window. WHAT I DO ☁️ Cloud Infrastructure — AWS and Azure environments built from scratch. Reusable Terraform modules across dev/test/prod. Least-privilege IAM, private networking, and security controls built in from day one rather than retrofitted before an audit. ☸️ Kubernetes & Delivery — EKS and AKS design, Helm charts, GitOps with Argo CD using App-of-Apps with auto-sync and self-healing. GitHub Actions pipelines with multi-arch builds and Trivy scanning inside the pipeline, not bolted on after. 📊 Observability — Full instrumentation, not just a dashboard. OpenTelemetry, distributed tracing with Jaeger, Prometheus with custom alert rules, Loki, and Grafana dashboards built for the questions your team actually asks at 3am. ⚙️ Backend & GenAI Pipelines — Python/FastAPI, Django, Node.js. Multi-tenant architecture with schema-per-tenant isolation. End-to-end LLM ingestion on Temporal — batch embedding, retry policies, vector databases — running unattended. 🌐 Networking & Production Firefighting — Routing protocols, VPC and Transit Gateway design, OpenVPN PKI, TLS termination, WAF, DNS, reverse proxies. When something is down and nobody knows why, this is the work I'm best at. 🚨 Backup, DR & Cross-Cloud Migration — Velero for cluster and persistent volume backups, database point-in-time recovery, geo-restore runbooks, and tested restore procedures. Live provider migration when it's needed. RESULTS - Migrated a client's full production workload from AWS to Azure in a single planned maintenance window with DNS-level fallback, during the March 2026 ME-CENTRAL-1 regional outage - Built complete backup and recovery for a production Kubernetes platform — Velero for cluster state and volumes, database PITR, and documented geo-restore runbooks, tested rather than assumed - Cut manual cloud operations ~95% with a serverless EC2 scheduling platform on Lambda, EventBridge Scheduler, and Terraform - Reduced order execution latency ~73% on a live automated trading system through Redis caching, async queue processing, and multi-broker integrations - Architected a multi-tenant SaaS platform on FastAPI and Keycloak SSO with schema-per-tenant PostgreSQL isolation and automated per-tenant provisioning - Reduced EKS attack surface ~90% with Calico micro-segmentation and Kyverno admission policies, and remediated SOC 2 findings surfaced by Vanta directly in Terraform - Eliminated ~90% of manual infrastructure setup and cut deployment time ~50% with standardized Terraform modules across AWS Organizations INDUSTRIES Fintech and algorithmic trading · Generative AI and document platforms · Healthcare (HIPAA identity infrastructure, 1M+ users) · Multi-tenant SaaS STACK Cloud: AWS, Azure · Terraform, Ansible Kubernetes: EKS, AKS, Helm, Argo CD, Istio, Calico, Velero CI/CD: GitHub Actions, GitLab CI, Jenkins Backend: Python, FastAPI, Django, Node.js, NestJS, Celery, Temporal Data: PostgreSQL, Redis, Elasticsearch, Qdrant Observability: OpenTelemetry, Prometheus, Grafana, Jaeger, Loki, EFK Security: Kyverno, Vanta, Trivy, SonarQube, Keycloak, WAF, OpenVPN Send me the problem rather than the job description — I'll tell you honestly whether I'm the right person for it, and what it will actually take.