Allocated students
Before no record uploaded

📘 About the Project

EduPath Admin is a smart university admissions platform designed to automate student placement based on secondary school performance. The idea was inspired by the repeated challenges students and administrators face during the admissions process: long delays, inconsistent allocations, and lack of transparency. I wanted to build a solution that could streamline decision-making while maintaining fairness and accountability.

🌱 Inspiration

During my community work and academic interactions, I noticed that many students struggle to secure placements in suitable programs because manual allocation systems often fail to account for student performance comprehensively. Administrators, on the other hand, spend countless hours processing Excel sheets and managing exceptions manually.

This motivated me to create a platform that:

Automates allocation
Provides clear reasoning for every assignment
Offers flexibility through manual overrides

📚 Learning Journey

Building EduPath Admin taught me:

Data normalization: Converting grades from multiple school systems into a unified scoring system
Allocation algorithms: Implementing a "best-fit" strategy where higher-score students are prioritized for higher-tier courses
Web development: Combining FastAPI for the backend and dynamic HTML/JS for the frontend with WebSocket support
PDF generation: Using WeasyPrint and Jinja2 templates to generate professional per-student and full allocation reports
User experience: Designing a demo-ready admin interface that is both intuitive and functional

🏗️ How We Built It

🔧 Backend

Python + FastAPI for API endpoints including allocation, manual override, and PDF generation

🎨 Frontend

HTML, CSS, and JavaScript with forms for uploading student Excel files, applying overrides, and downloading PDFs

📁 Data Handling

universities.json contains all universities and their programs
pandas reads student Excel files and converts letter grades to numeric scores

🧮 Allocation Logic

Each student is scored based on their grades
Courses are assigned using a best-fit algorithm that ensures fairness and maximizes student-course alignment
Overrides can be applied manually and reflect instantly in the UI

📄 Reporting

Each student receives an HTML report with allocation reasoning
Full allocations are exported as multi-page, styled PDFs

🛠️ How It Works

1. Upload Student Records

Admins upload an Excel file containing students’ grades.

2. Data Normalization

Grades from multiple systems (A/B/C, numeric) are converted into a unified score.

3. Allocation Algorithm

For each student ( i ) with total score ( S_i ), the algorithm evaluates all courses ( c ) across universities.

The best-fit course ( c^* ) is chosen such that:

$$ c^* = \arg\max_{c \in \text{available courses}} \text{ScoreAlignment}(S_i, c_{\text{min_score}}) $$

Higher-score students get priority for higher-tier courses.

4. Manual Overrides

Admins can adjust allocations instantly; changes reflect immediately in the frontend.

5. Reporting

HTML and multi-page PDFs are generated automatically, including detailed allocation reasoning.

6. Download

Admins can download per-student PDFs or a full allocation PDF.

🔁 Flowchart

Excel Upload → Grade Normalization → Best-Fit Allocation → HTML Report
                                                    ↓
                                              PDF Generation
                                                    ↓
                                        Manual Override (if needed)

⚠️ Challenges Faced

Grade Normalization: Different grading systems (A/B/C vs numeric) required a robust conversion system
Dynamic Allocations: Ensuring real-time updates in the frontend after overrides required implementing live-refresh strategies
PDF Styling: Creating professional, multi-page PDFs that were responsive and included 20+ students per page without breaking formatting
Data Structure: Switching from nested country-based JSON to a flat university list simplified processing while maintaining flexibility

🚀 Future Plan for Expansion

Database Integration: Store students, universities, and allocations persistently for large-scale deployments
Analytics Dashboard: Visualize allocation trends, student success rates, and bottlenecks
Authentication & Roles: Different access levels for admins, reviewers, and regional coordinators
AI-Assisted Recommendations: Suggest alternative courses for borderline students using predictive modeling
Multi-Language & Accessibility: Expand the platform for diverse regions with localization and accessibility features