BENNY MALITI

🎯 Challenge: Build production-grade, highly available multi-tier e-commerce platform on Amazon EKS
🛠️ Key Tech: Terraform, Amazon EKS, Kubernetes, Helm, Docker, RDS MySQL, ElastiCache Redis, React, Node.js, Prometheus, Grafana
✨ Outcome: End-to-end Kubernetes orchestration with container security, observability, and automated CI/CD

• Provisioned 76 AWS resources using Terraform IaC including EKS cluster, RDS MySQL, ElastiCache Redis, ECR, VPC with public/private subnets across 3 AZs
• Configured IAM Roles for Service Accounts (IRSA) for least-privilege pod security
• Deployed React/Nginx frontend and Node.js backend API as Kubernetes Deployments with HPA autoscaling, health probes, and rolling update strategy
• Routed traffic via AWS Load Balancer Controller Ingress
• Containerised all services using multi-stage Docker builds pushed to ECR; configured Kubernetes Secrets and ConfigMaps
• Deployed full observability stack using Helm: Prometheus for metrics scraping and Grafana for dashboards
• Built 3-stage GitHub Actions CI/CD pipeline (Test → Build & Push to ECR → Deploy to EKS) with automated smoke tests and zero-downtime rolling deployments

🎯 Challenge: Build production-grade multi-region disaster recovery infrastructure achieving 15-minute RTO and 1-hour RPO
🛠️ Key Tech: Terraform, AWS Aurora MySQL, ECS Fargate, S3, WAF, GuardDuty, AWS Backup, Docker, GitHub Actions
✨ Outcome: Fully automated DR infrastructure with cross-region replication and enterprise security controls

• Architected multi-region DR across London & Ireland using 11 modular Terraform modules
• Implemented Aurora MySQL cross-region replication with automatic failover capabilities achieving Pilot Light DR strategy
• Deployed containerised Node.js application on ECS Fargate with Application Load Balancer and auto-scaling
• Configured S3 cross-region replication with KMS encryption for healthcare data compliance
• Implemented WAF with OWASP Top 10 protection rules and GuardDuty threat detection for enterprise security
• Built 3 CI/CD pipelines: Terraform validation, container builds with Trivy scanning, DR validation testing
• Achieved 15-minute RTO and 1-hour RPO disaster recovery targets with documented runbooks
• Set up AWS Backup with cross-region vault copies and 35-day retention for compliance requirements

🎯 Challenge: Build production-grade real-time serverless analytics platform processing Reddit data through a 7-Lambda microservices pipeline
🛠️ Key Tech: Terraform, Python 3.12, AWS Lambda, Kinesis, DynamoDB, Amazon Comprehend, API Gateway, S3, GitHub Actions
✨ Outcome: Processing 100+ posts per run with NLP sentiment analysis, engagement scoring, and real-time streaming

• Architected 7 Lambda microservices (Python 3.12, ARM64) spanning data ingestion → Kinesis streaming → parallel analytics fan-out → DynamoDB storage — processing 100+ posts per run
• Implemented real-time NLP using Amazon Comprehend for sentiment analysis, key phrase extraction, and entity detection across 7 subreddits
• Built 8 modular Terraform IaC modules managing 17 AWS services with S3 remote state, DynamoDB locking, and KMS encryption
• Engineered 4-stage GitHub Actions CI/CD pipeline: Python syntax + Black + Flake8 + pytest (unit & integration with moto mocking), Terraform validate + plan on PRs, auto-deploy on main merge
• Implemented 5-table DynamoDB schema with GSIs for engagement scoring, sentiment querying, and trending topic detection
• Added engagement scoring algorithm (composite score 0–100), comment velocity calculation, and bot likelihood detection for author profiling
• Configured full observability stack: CloudWatch dashboards, SNS alerts, SQS dead-letter queues, and AWS X-Ray distributed tracing