3rd-year CS student. Love building backend systems and cloud tools to help a fast-paced team grow.

medium

github

kaushiik arul  • Peerlist

Built a serverless AWS automation tool to detect and clean up unused cloud resources, improving infrastructure hygiene and reducing cloud costs using AWS Lambda, EC2, EBS, IAM, CloudWatch, EventBridge, and Python (Boto3).

CloudOverwatch – Serverless AWS Resource Clea

Built a machine learning–based analytics system to evaluate DSA performance and generate personalized learning insights using Python, incorporating data preprocessing, feature engineering, and model training techniques including classification (Naive Bayes), clustering (K-Means), and association rule mining (Apriori).

CodeStruct – Machine Learning Model Training 

Built and deployed a production-style backend API called JobFlow using AWS cloud infrastructure and automated CI/CD pipelines. The application is a NestJS REST API containerized with Docker and deployed on Amazon ECS Fargate, with traffic routed through an Application Load Balancer and DNS managed by Amazon Route 53. The backend connects to a MySQL database hosted on Amazon RDS, which runs inside private subnets for security. The entire infrastructure was provisioned using Terraform (Infrastructure as Code), including the VPC, subnets, security groups, ECS cluster, load balancer, and database. A GitHub Actions CI/CD pipeline automatically builds Docker images, pushes them to Amazon ECR, and triggers ECS deployments on every code update. This architecture follows common production patterns such as private subnet isolation, load-balanced container services, automated deployments, and infrastructure version control.

Production-Ready Backend Deployment on AWS (E

Designed a scalable search system using TypeSense and Elasticsearch, implementing efficient indexing and query optimization to enable fast, reliable, and accurate search across large datasets.