An AI-powered recruitment automation system built by DuelBot (Hrishikesh & Rohit) for Accenture Hackathon 2025.
- Automatically parses Job Descriptions (JDs) using NLP
- Extracts and processes resumes (CVs) from PDFs
- Calculates AI-based match percentage between JDs and CVs
- Filters and shortlists candidates based on match threshold (e.g. ≥ 80%)
- Schedules interviews for shortlisted candidates
- Sends personalized interview emails
- Interactive Tkinter-based GUI dashboard
- Real-time status, logs, and countdown timer
GetSelect-AI uses a modular multi-agent design, each responsible for a specific task:
| Module | Responsibility |
|---|---|
covertpdf.py |
Converts resumes from PDF format to raw text and extracts structured info |
JDParsing.py |
Summarizes JDs and extracts key components like skills, qualifications, etc. |
newmatric.py |
Matches extracted resume data with JD and calculates match percentage |
advanced_matching.py |
Stores shortlisted candidates in SQLite and handles advanced matching logic |
schedular.py |
Generates interview slots and schedules interviews for selected candidates |
interviewmail.py |
Sends interview confirmation emails to candidates with time, date & details |
main.py |
One-click GUI to control all steps with visual logs and interview timer |
Install all dependencies via:
pip install -r requirements.txt