AI-Driven Candidate Assessment & ATS Platform

Full Stack DeveloperMar 2025 – Present · iFocus Systec

Overview

Qwikhire is an online AI-driven ATS platform built to help companies shortlist the right candidates through automated assessments. It conducts a structured four-round test portal covering General, Position-based, Coding, and Aptitude rounds, all powered by AI. A complete proctoring system monitors candidates in real time, detecting suspicious activity such as multiple faces or screen recording. After each test, the platform generates a detailed AI report analyzing every question and answer with insights into strengths, weaknesses, and overall performance, enabling companies to make confident hiring decisions without manual effort.

What I Built

  • Implemented role-based login with JWT authentication, refresh token rotation, and CSRF token protection across all user roles
  • Built on a microservice architecture with an API Gateway as the single entry point routing requests to independent backend services
  • Developed the complete candidate test portal with 4 rounds: General, Position, Coding, and Aptitude with AI-powered evaluation
  • Integrated OpenAI and FastAPI for AI-driven question generation, answer analysis, and assessment scoring across all rounds
  • Integrated AssemblyAI for speech-to-text and OpenAI TTS for text-to-speech in the conversational interview round
  • Implemented Judge0 for live coding round execution with real-time test case validation across multiple languages
  • Built complete AI candidate report with per-question analysis covering strengths, weaknesses, and overall performance
  • Implemented proctoring with suspicious activity detection including no face, multiple faces, screen recording, and camera capture stored on GCP
  • Implemented Redis for backend session management and fast token state handling across microservices
  • Built a queue system for report generation so concurrent report requests are processed sequentially without conflicts
  • Assisted in deployment and creation of GCP Cloud Run services for scalable cloud infrastructure

Tech Stack

Frontend

React JSTailwind CSS

Backend

JavaSpring BootSpring Security
MicroservicesFastAPIPython

AI & Integrations

OpenAIAssemblyAIJudge0

Database & Cache

RedisMySQLMongoDB

DevOps

GCP Cloud RunGitHub

Features

AI-Driven Candidate Assessment

Four-round test portal covering General, Position-based, Coding, and Aptitude rounds, fully automated and AI-evaluated.

ATS & Candidate Shortlisting

Automatically shortlists candidates based on their test performance and AI score, removing manual screening overhead.

Complete Proctoring System

Real-time detection of suspicious activity including no face, multiple faces, screen recording, and camera capture stored on GCP.

AI Report Analysis

Detailed per-question AI analysis of candidate answers covering strengths, weaknesses, and overall performance metrics.

Microservice Architecture

Gateway server acts as the single entry point routing requests to independent backend services for scalability and isolation.

Queue-Based Report Generation

Report generation runs through a queue system so concurrent report requests are processed one at a time without conflicts.

Session & Redis Management

Backend sessions maintained using Redis for fast, reliable token and session state management across microservices.

Coding Round with Judge0

Live coding assessment execution powered by Judge0, supporting multiple languages with real-time test case validation.