# 🎓 FastAPI Backend for AI-Powered University Admission Management

We want to build a **complete, scalable backend application with FastAPI** to power a next-generation **student admission platform**.

The goal is to enable **universities** to publish admission projects (programs or seats available) and manage the admission process efficiently, while **students** interact through a mobile platform. **Gemini AI** will be integrated to accelerate workflows, generate insights, and assist in decision-making.

---

## 🚀 Core Functional Modules

1. **Admission Projects Management (Universities)**

   * Create new admission projects via an AI-assisted conversational interface.
   * Define selection criteria (skills, academic requirements, soft skills).
   * Display and manage projects in a dashboard with search and filtering.

2. **Student Applications Database**

   * List and filter all student applications.
   * Advanced search by status, criteria, or program.
   * Bulk actions (update status, send messages).
   * Export student data.
   * Track required documents (CV, transcripts, certificates).

3. **Detailed Student Profile**

   * Centralized view of student data and uploaded files.
   * Manage academic history and personal information.
   * Internal notes for admission staff.
   * Full communication history.
   * AI-generated evaluations of student fit for a program.

4. **Communication**

   * Send messages (email / in-app) to one or multiple students.
   * AI-powered drafting and optimization of communications.
   * Message templates library.
   * Track all communications.

5. **Reports & Analytics (AI-powered)**

   * Dashboards with statistics (applications, acceptance rates, etc.).
   * Student ranking based on weighted criteria.
   * Visualizations (radar charts, distributions, trends).
   * AI insights on efficiency of admissions.

6. **Authentication & Users**

   * University accounts with secure authentication (JWT).
   * Role-based access (admin, staff).

---

## 🔗 Proposed FastAPI Endpoints

* **Auth** (`/auth`): login, JWT token, current user.
* **Projects** (`/projects`): CRUD for admission projects.
* **Students** (`/candidates`): CRUD for applications, bulk actions.
* **Documents** (`/candidates/{id}/documents`): upload, list, download.
* **Notes** (`/candidates/{id}/notes`): internal notes.
* **Communications** (`/communications`): send messages, templates, history.
* **AI Services** (`/ai`): project creation assistant, candidate assessment, message composition.
* **Reports** (`/reports`): candidate rankings, statistics dashboards.

---

## 📦 Backend Entity Model

Below is the **data model (ERD)** that structures the application.

### 1. **User (University Staff)**

Represents a university staff member who manages projects and candidates.

* id (PK), email (unique), hashed\_password, full\_name, created\_at
* Relations: creates many Projects, writes many Notes

### 2. **Project (Admission Project)**

Represents an admission campaign or available program.

* id (PK), title, description, status (`draft | open | closed`), created\_at, owner\_id (FK → User)
* Relations: belongs to User, has many Criteria, linked to many Candidates, linked to many SoftSkills

### 3. **Candidate (Student Applicant)**

Represents a student applying to one or multiple projects.

* id (PK), first\_name, last\_name, email (unique), phone, status\_overall (`active | archived | blacklisted`)
* Relations: many Projects (via Applications), many Documents, many Notes, many Communications, many AIAssessments

### 4. **Application (Student ↔ Project Association)**

Pivot table linking a Candidate to a Project.

* candidate\_id (FK), project\_id (FK), status (`new | in_review | interview | rejected | admitted`), applied\_at

### 5. **Document**

Represents uploaded files (CV, transcript, etc.).

* id (PK), candidate\_id (FK), file\_name, file\_path, document\_type (`cv | transcript | cover_letter | other`), uploaded\_at

### 6. **Note**

Internal note written by university staff about a student.

* id (PK), candidate\_id (FK), user\_id (FK), content, created\_at

### 7. **Communication**

Represents a message sent to a candidate.

* id (PK), candidate\_id (FK), user\_id (FK, nullable), subject, body, type (`email | sms | platform`), sent\_at

### 8. **AIAssessment**

AI evaluation of a student for a specific project.

* id (PK), application\_id (FK), score (0–100), summary, details (JSON), generated\_at

### 9. **Criterion (Selection Criterion)**

Represents a required criterion for a project.

* id (PK), project\_id (FK), description, weight (float, default 1.0)

### 10. **SoftSkill**

Represents a non-technical skill (e.g., Communication, Teamwork).

* id (PK), name (unique)
* Relations: many Projects, connected to Candidate assessments

---

## 🏗️ Tech & Scalability Requirements

* **FastAPI** backend (async, type-safe, high performance).
* **PostgreSQL** as relational database.
* **Docker + CI/CD** for deployment.
* **Redis / Celery** for background tasks (bulk processing, AI evaluations).
* **Pydantic models** for validation.
* **Secure authentication** with JWT & role-based permissions.
* **Gemini AI integration** for admission project creation, candidate assessments, and communication drafting.

---

⚡ **Final Objective**:
Deliver a **production-ready FastAPI backend** that empowers **universities** to streamline and accelerate student admissions with **AI-powered automation**, serving both a **web platform for universities** and a **mobile app for students**.