HireDesk

Home page for the app after login/signup
ML Score ranking for an applied candidate
Interview room interface via WebRTC
Recruiter Dashboard for in depth analysis
Inbuilt chatbot for queries
Jobs page on the candidate side

Inspiration

Hiring today feels broken, especially for startups. Recruiters receive hundreds of resumes for a single role and spend hours manually screening them. At the same time, candidates don’t know how strong their resumes actually are or why they get rejected. With the rise of remote interviews and AI tools, another challenge has emerged - maintaining integrity during online interviews. Recruiters often worry about candidates switching tabs, reading from another screen, or getting outside help. We built HireDesk to make hiring smarter and more transparent. The idea was simple - combine AI-powered resume analysis with an in-platform interview system that gives recruiters better visibility, while also helping candidates improve their resumes.

What it does

HireDesk is an AI-powered hiring platform that supports both recruiters and candidates. For candidates:

Upload resume in PDF or DOCX format
Get a resume score based on the job description
Receive suggestions to improve skills and keywords
Use an AI assistant to ask doubts about their resume
Apply to jobs through a simple dashboard

For recruiters:

Post job listings with detailed role descriptions
Upload resumes individually
Upload resumes in bulk using CSV files
Automatically rank candidates using AI-based resume scoring
Compare candidates based on resume scores
Conduct interviews directly within the platform
Monitor interviews with face presence detection
Detect prolonged face absence
Detect multiple faces in the camera frame
Detect tab or window switching during interview
Transcript generation during interviews ( same as Google Meet )
Access recruiter dashboard with analytics and activity logs (Recommended to see the demo yt video for understanding flow and trying the product).

How we built it

Built the frontend using React and Tailwind CSS to create a clean and responsive interface.
Implemented role-based authentication using Firebase with email verification and Google login.
Developed the backend using Django and Django REST Framework for APIs and business logic.
Integrated resume parsing using spaCy for extracting structured information from resumes.
Designed a semantic similarity-based scoring system to match resumes with job descriptions.
Implemented AI-based resume improvement suggestions for candidates.
Built the in-platform interview system for conducting interviews directly inside HireDesk.
Developed face detection and monitoring logic using computer vision techniques.
Implemented tab switching detection using browser APIs.
Used PostgreSQL for structured and secure data storage.
Deployed Frontend, Backend and ML services on DigitalOcean.
Configured NGINX and environment variables to ensure secure production deployment.

Challenges we ran into

Designing a fair resume scoring system without relying only on keyword matching
Handling resumes in different formats such as PDF and DOCX consistently
Reducing false alerts in face detection when candidates slightly moved away from the camera
Managing real-time interview monitoring without affecting system performance
Preventing fake user registrations by implementing proper authentication flows
Integrating ML services smoothly with backend APIs for real-time dashboard updates

Accomplishments that we're proud of

Built a fully functional AI recruitment platform from scratch
Successfully combined resume intelligence and interview monitoring in one system
Implemented bulk resume upload with automated candidate ranking
Created a structured interview alert timeline for recruiters
Deployed the entire system on cloud infrastructure with secure authentication
Ensured proper separation of frontend, backend, and ML components for scalability