GDGC AI Applicant Evaluation System

An AI-powered applicant screening tool built by the GDG on Campus University of Jeddah tech team. This system automates the evaluation and ranking of club membership applicants using configurable scoring criteria and AI analysis — replacing manual review with a consistent, data-driven pipeline.

---

How It Works

The system processes applicant data from a Google Sheet or uploaded Excel/CSV file through three stages:

1. Eligibility Check — AI reads each applicant's motivation text and checks it against custom criteria defined by the organizer (e.g. "technical background", "AI interest"). Each applicant is marked Accepted or Rejected.

2. Column Scoring — The organizer selects which columns to score (e.g. motivation statement, project experience) and assigns a weight to each. The AI scores every selected field from 0–100 based on clarity, specificity, passion, and relevance to the event.

3. Final Ranking — A weighted total score is calculated for each applicant and they are ranked from highest to lowest.

---

Scoring System

Eligibility is always fixed at 30% of the total score. The remaining 70% is distributed across the columns the organizer selects, with custom weights that must sum to 100%.

Component	Weight
Eligibility (Accepted = 100, Rejected = 0)	Fixed at 30%
Organizer-selected columns (AI-scored 0–100)	70% total, split by custom weights

Formula:

Total Score = (Column Scores × their weights × 0.70) + (Eligibility Score × 0.30)

---

Project Structure

GDG-Automation-Project/
├── main.py               # Core pipeline: data loading, eligibility, scoring
├── app.py                # Streamlit admin UI
├── requirements.txt      # Python dependencies
├── .gitignore
└── assets/               # UI images and logos

---

Setup

1. Clone the repository

git clone https://github.com/gdguj/GDG-Automation-Project.git
cd GDG-Automation-Project

2. Install dependencies

pip install -r requirements.txt

3. Add your API key

Create a .env file in the project root:

GROQ_API_KEY=gsk_xxxxxxxxxxxxxxxxx

4. Run

Terminal mode (interactive):

python main.py

Streamlit admin panel:

streamlit run app.py

---

Using the Admin Panel

1. Enter the event name
2. Load data via Google Sheet URL or upload an Excel/CSV file
3. Select which column contains the motivation/eligibility text
4. Choose which columns the AI should score
5. Set weights for each scoring column (must total 100%)
6. Enter eligibility criteria (one per line)
7. Click Run Evaluation

Results are displayed in a ranked table and can be downloaded as a CSV.

---

Output

The pipeline produces a results CSV with:

Column	Description
[Column]_Score	AI score for each selected column (0–100)
[Column]_Feedback	AI feedback for each scored column
Eligibility	Accepted or Rejected
Eligibility_Score	100 if Accepted, 0 if Rejected
Total_Score	Final weighted score (0–100)
Rank	Applicant ranking (1 = best)

---

Tech Stack

• Python — core pipeline
• Groq API (LLaMA 3.1) — AI scoring and eligibility analysis
• Pandas — data processing
• Streamlit — admin UI
• Google Sheets / Excel / CSV — applicant data sources

---

Team

Developed by the GDG on Campus University of Jeddah development team — 2025/2026.