KaamAI (Cloud Based Resume Screening)

“It is a cloud-based artificial intelligence system designed to automate resume screening and candidate ranking for recruitment organizations.”

01. What the Project Is

KaamAI is a cloud-native resume parsing and ranking platform engineered as the final capstone showcase for my graduate-level Distributed Cloud Computing course. The name cleverly synthesizes the Nepali word 'Kaam' (meaning Job) with 'AI' (Artificial Intelligence) to create a linguistic double-entendre that translates to 'Earning' ('Kaamai') in Nepali. The system functions as a highly scalable enterprise human-resources engine designed to automate talent ingestion and rank candidate profiles using automated text analytics pipelines.

02. Why Technical Choices Were Made

The primary architectural constraint for our graduate project was designing a system capable of running on distributed cloud topology while optimizing infrastructure budgets. We chose AWS because its academic free-tier allowed us to experiment with production-grade scaling features without generating operational overhead. Rather than deploying a basic static dashboard, we made the strategic decision to integrate a custom machine learning model directly into our ingestion queues to solve the problem of high-volume candidate screening friction.

03. How It Was Implemented

The platform is divided into a decoupled Node.js backend controller, a highly responsive React frontend, and an isolated Python microservice that handles semantic analysis calculations for candidate scoring. For infrastructure, we containerized our application services using Docker. I then configured a robust CI/CD pipeline using GitHub Actions; every verified commit automatically builds a production-ready image, pushes the binary to AWS Elastic Container Registry (ECR), and orchestrates an automated deployment directly onto live EC2 container runtimes.

> stream closed // text layer read complete.