An end-to-end Workforce Intelligence Platform that predicts talent shortages, hiring timelines, and skill gaps by connecting industry trends, company behavior, and student readiness.
This project goes beyond dashboards — it turns workforce data into actionable decisions for both industries and job seekers.
Industries struggle to answer:
- When will we need to hire?
- Is our talent pipeline strong enough?
- Which companies will face hiring pressure first?
Students struggle to answer:
- Which industries are hiring next?
- Which companies should I target?
- How ready is my resume for future jobs?
This system bridges that gap.
The platform uses rule-based, explainable AI logic (not black-box models) to:
- Forecast workforce risk
- Predict hiring surge timelines
- Compare companies within an industry
- Guide students on skills, jobs, and resume readiness
Same data. Different decisions.
- Talent Supply vs Demand analysis
- Workforce Risk Score (0–100)
- Dynamic baseline risk
- Hiring Surge Prediction (2026 only)
- What-If Simulation for workforce planning
- Hiring outlook in simple terms
- Competition level (High / Balanced / Opportunity-rich)
- Skill demand intelligence
- Career preparation guidance
- Industry switch suggestions
- Compare companies within the same industry
- Supply, Demand, Risk & Hiring Timeline
- Synthetic but logically derived company data
- ATS Match Score (0–100)
- Skill gap detection:
- Critical gaps
- Industry alignment gaps
- Future readiness gaps
- Resume improvement guidance
- Aligned with future hiring trends
- JWT-based authentication
- Two roles:
- INDUSTRY_USER → Industry Dashboard only
- STUDENT_USER → Student Dashboard, Company Comparison, Resume Analyzer
User (Browser) ↓ Frontend (React + TypeScript + Tailwind) ↓ REST APIs (JSON) Backend (FastAPI – Python) ↓ Analytics & Simulation Logic ↓ SQLite Database (Dev) / PostgreSQL (Prod-ready)
Supply = Internship_Intake × Conversion_Rate
Demand = Growth_Rate + (Attrition_Rate × 1.5)
- Percentile-based normalization (P5–P95)
- Applied per industry
- Prevents artificial 0/100 spikes
Core_Risk = (Demand_Score − Supply_Score) + (Attrition × 15)
Baseline_Risk = 5 + (Attrition × 10) + (Demand_Trend × 0.5)
Final_Risk = max(Core_Risk, Baseline_Risk)
HPI = (Demand − Supply) + (Attrition × 20) + (Demand_Trend × 0.8)
Mapped to:
- 1–3 months → Immediate hiring
- 4–6 months → Near-term hiring
- 6–12 months → Planned hiring
- React + TypeScript
- Tailwind CSS
- Recharts
- FastAPI (Python)
- Pydantic
- SQLAlchemy
- SQLite (development)
- PostgreSQL (production-ready)
- JWT Authentication
- Role-Based Access Control
- Password hashing (bcrypt)
- Synthetic but derived data
- No random values at runtime
- Company data derived from industry baselines
- Deterministic and reproducible outputs
- Percentile normalization avoids misleading extremes
- Debug logging for intermediate values
- Same inputs → same outputs
- Stable across refreshes
- Transparent and judge-friendly
🔹 Backend Setup
cd backend
pip install -r requirements.txt
uvicorn api:app --reload
🔹 Frontend Setup
cd frontend
npm install
npm run dev
Unlike generic dashboards, this project focuses on high-impact insights:
Specific Utility: Not a LinkedIn clone or a generic ML dashboard.
Predictive Power: Focuses specifically on when hiring will happen.
Bridging the Gap: Directly connects industry decisions with student career planning.
Transparency: Built with Fully Explainable AI (XAI) logic so users understand the "why" behind the predictions.
-
Real Job Postings: Integration with live job boards.
-
Resume Versioning: Track how different resume iterations perform.
-
Skill Similarity Mapping: Visualizing how current skills align with market demand.
-
Admin Analytics: Insights for institutional or platform administrators.
-
Cloud Deployment: Moving from local hosting to AWS/GCP/Azure.
This project is for educational and demonstration purposes.
Keshav Agarwal