Building production-grade infrastructure with Kubernetes, AWS, and GitOps
I'm a B.Tech CSE graduate (2026) from Roorkee Institute of Technology who has spent the last two years building production-grade DevOps systems from scratch — not tutorials, but real multi-service platforms with CI/CD, GitOps, observability stacks, and chaos engineering.
- 🔭 Currently building on AWS EKS · Kubernetes · Terraform · ArgoCD
- 🛡️ Hall of Fame @ Axway.com for responsible disclosure of a critical source code exposure vulnerability
- 📊 Reduced MTTR from 45 min → 8 min on a distributed tracing platform
- ⚡ Provisioned 11-microservice infra in 12 minutes with Terraform (down from 4 hours)
- 🎯 Preparing for AWS SAA-C03 and Terraform Associate certifications
- 💬 Ask me about Kubernetes failure modes, SLO burn-rate alerting, or GitOps pipelines
Cloud & Infrastructure
Containers & Orchestration
CI/CD & GitOps
Observability
Security & DevSecOps
Languages & Scripting
Spring Boot · Jenkins · ArgoCD · Terraform · Karpenter · Istio · Kyverno · Velero · Prometheus · Grafana · Loki · Tempo
- Architected full CI/CD + GitOps pipeline on EKS; provisioned all infra via modular Terraform with S3/DynamoDB state backend
- Blue-green deployments via Argo Rollouts with zero-downtime releases; Karpenter for dynamic node autoscaling
- DevSecOps pipeline: Trivy + OWASP ZAP + SonarQube on every push; Kyverno admission policies + Istio mTLS
- Full observability: Prometheus/Grafana/Loki/Tempo; Velero + MinIO 6-hour DR backups targeting 99.9% availability
Prometheus · Grafana · Alertmanager · OpenTelemetry · Jaeger · Tempo · Loki · ArgoCD · Sloth · Chaos Mesh · ELK · FastAPI
- OpenTelemetry auto-instrumentation across 3 FastAPI microservices (5 signal types: metrics, logs, traces, spans, exemplars)
- MTTR reduced from 45 min → 8 min via distributed tracing and log correlation
- Sloth-based SLO/SLI: 32 Prometheus recording rules, multi-burn-rate alerting (14.4×/6×), zero SLO breaches over 90 days; false positives cut 67%
- Chaos engineering via Chaos Mesh (pod kill, network latency, CPU stress); alert firing confirmed within 90 sec of fault injection
- ArgoCD App-of-Apps GitOps: deploy time 30 min → 90 sec, zero config drift
AWS EKS · Jenkins · ArgoCD · Terraform · Helm · ECR · Route53 · ACM · Trivy · Redis · gRPC · Secrets Manager
- Deployed cloud-native 11-microservice platform (Java, Go, Python, Node.js, gRPC) on EKS
- Terraform automation reduced infra provisioning from 4 hours → 12 minutes (VPC, IAM, 11 ECR repos)
- 11 parallel Jenkins pipelines with Trivy scanning; end-to-end deploys under 5 min across dev/staging/prod
- 40+ vulnerabilities caught pre-production; zero security incidents over 6 months
| 🏅 Hall of Fame — Axway.com | Responsibly disclosed a critical vulnerability exposing platform source code with full exploitation chain documentation |
| 🔐 Security Research | Disclosed critical vulnerabilities (auth bypass, price tampering, data exposure) across 5+ organizations with 100% remediation rate |
I'm actively looking for my first DevOps / Cloud / SRE role at funded startups (Series A–C) or any other organization. If you're building something interesting, reach out.
