TalentMatchAI 🎯
Revolutionizing hiring with AI fairness, speed, and transparency.
AI-Powered Fair Hiring Platform for SMEs, Corporates, and Government Agencies
🚀 WHY TALENT MATCH I MATTERS
Traditional hiring is costly, biased, and time-consuming.
SMEs often lack HR capacity to screen hundreds of applicants.
Governments require transparent, accountable recruitment.
Corporates must maintain diversity, inclusion, and efficiency.
👉 TalentMatchAI transforms this process by automating fair, explainable, and data-driven applicant ranking.
💡 Core Value Proposition
📂 Automated CV Parsing – Extracts data from PDF/DOCX instantly.
⚖ Fairness-Driven Scoring – Every applicant gets objective, explainable rankings.
🤝 Transparency for Trust – Each score includes reasons and weights.
📊 Data-Driven Insights – Improves hiring outcomes & retention rates.
🧠 TalentMatchAI is beyond a hackathon project — it’s designed for real HR and government adoption.
✨ KEY FEATURES
Backend (FastAPI)
RESTful API for applicant uploads, parsing, and ranking.
Machine learning scoring model using scikit-learn and pandas.
JSON output with detailed reasoning for every rank.
Modular structure for easy scaling into HR or government systems.
Frontend (HTML / CSS / JavaScript)
Clean, responsive interface for HR teams and SMEs.
Input form for applicant data (skills, education, experience).
Displays ranked results instantly with reasons.
Connects directly to the FastAPI backend.
🧱 FULL PROJECT STRUCTURE
TalentMatchAI/ ├── backend/ │ ├── app/ │ │ ├── main.py # FastAPI entry point │ │ ├── models/ # Applicant data models │ │ ├── routers/ # Upload & ranking endpoints │ │ ├── services/ # Parsing & scoring logic │ │ └── utils/ # Helper scoring functions │ ├── datasets/ │ │ └── applicants.json # Sample dataset for testing │ ├── requirements.txt # Backend dependencies │ └── README.md │ ├── frontend/ │ ├── index.html # Main user interface │ ├── style.css # Styling and responsiveness │ ├── script.js # API integration and interactivity │ ├── docs/ │ └── pitch-deck-outline.md # For investors, agencies, and partners │ ├── LICENSE └── .gitignore
⚙️ Backend Setup Instructions
- Clone the repository
git clone https://github.com/Agrictechventure68/TalentMatchAI.git cd TalentMatchAI/backend
- Create and activate virtual environment
python -m venv venv venv\Scripts\activate # On Windows source venv/bin/activate # On Mac/Linux
- Install dependencies
pip install -r requirements.txt
- Run the backend server
uvicorn app.main:app --reload
✅ Server Live At: http://127.0.0.1:8000
📡 API ENDPOINTS
Method Endpoint Description
POST /upload/ Upload applicant CV (PDF/DOCX) GET /rank/ Retrieve ranked applicant list
Example Response:
[ { "name": "John Ade", "score": 87, "reasons": ["5 years experience", "Python & SQL skills"] }, { "name": "Victoria Ifijeh", "score": 72, "reasons": ["Strong education", "Limited direct experience"] } ]
🧩 Fairness Guarantee
No hidden algorithmic weights.
All reasoning and scoring criteria are visible.
Configurable scoring ensures fairness across gender, background, and academic pedigree.
🌐 Frontend (Demo Interface)
1️⃣ Open the Frontend
Go to the frontend/ folder
Open index.html in your browser (or use VS Code Live Server)
2️⃣ Add Applicants
Enter multiple applicants with:
Name
Skills (comma separated)
Education
Years of Experience
3️⃣ RANK APPLICANTS
Click “Rank Applicants” → The page calls your backend API and displays ranked results with fairness reasoning.
Example Screenshot (Frontend View):
Name Score Reason
Alice 95 Excellent skills, strong experience John 89 Balanced background Mary 78 Moderate experience
This interface can connect to a local FastAPI backend or a deployed Render/Cloud backend at https://talentmatchai.onrender.com/rank-applicants
💼 IMPACT FOR ORGANIZATIONS
Stakeholder Benefit
SMEs Save up to 70% HR screening time Corporates Reduce hiring bias & increase retention Government Agencies Achieve transparency and public trust
🔮 FUTURE ROADMAP
🗂 NLP-powered CV parsing (with spaCy or Transformers)
🎯 Job-role–specific ranking models
📊 Admin dashboards for HR visualization
🌍 API deployment on Render / AWS / GCP
🤖 AI-driven interview question generator
👤 AUTHOR
Bright Doro (Agrictechventure68) Educator | Agricultural & Tech Consultant | Software Developer
Mission: Build fair, efficient, and impactful tech for Africa’s growth.
🌐 GitHub: Agrictechventure68 📧 agriempower4dcentury@gmail.com
📜 LICENSE
This project is licensed under the MIT License — free to use, modify, and distribute with credit.