Projects

AI systems, RAG pipelines, and full-stack products I've built along the way.

Loadbalancer for AWS – Multi-Tier Infrastructure

A production-grade, highly available three-tier AWS infrastructure fully provisioned with Terraform. Web, App & Database tiers, Auto Scaling, Application Load Balancers, and Multi-AZ RDS out of the box.

Auto ScalingRDS

CycloCast – AI Cyclone Intelligence Platform

Built an AI-powered tropical cyclone tracking platform for IMD/MoES (SIH 2026) that fuses multi-source satellite imagery with deep learning to deliver real-time storm detection, IMD classification, 72-hour path forecasts with uncertainty cones, and historical replay — processing Meteosat IR telemetry and GraphCast atmospheric data.

Full-Stack Enterprise ERP System

Engineered a production-ready Full Stack ERP application to automate manufacturing lifecycles, processing over 1,000 inventory items and securing production tracking operations.

CI/CD

Render

TerraStorm – Cloud-Native AI Platform

Architected a Cloud AI platform utilizing an event-driven AWS serverless architecture to automate land-use change detection across 10,000+ square kilometers of satellite imagery.

AWS ServerlessDjangoMachine LearningS3LambdaSageMakerSNS

Brains – Highly Available Cloud Architecture

Designed and deployed a production-grade, highly available web architecture on AWS with multi-AZ auto-scaling (2–4 EC2 instances), Application Load Balancer, Dockerized microservices, Terraform IaC (20+ resources), and CI/CD automation — built as a case study with experimental verification of failover, scaling, and rollback.