Internship Management System (IMS)

A multi-application platform for managing student internships, connecting students, companies, and faculty through a shared backend. Built as a university project for the Internet Programming course at the Faculty of Electrical Engineering, University of Banja Luka.

Overview

The system supports the full internship lifecycle: students build a CV profile and apply to internship postings, companies review applications and manage ongoing internships, and faculty oversee the process, evaluate outcomes, and manage company/student records. An AI-powered recommendation engine suggests relevant internships to students based on their profile.

Three separate client applications talk to a single REST API backend, each tailored to a specific user role:

Application	Role	Stack	Port
backend	Shared REST API	Spring Boot, PostgreSQL	8080
student-app	Students	React + TypeScript (Vite)	5173
company-app	Companies	Java Servlets + JSP (embedded Tomcat)	8081
faculty-app	Faculty staff	Java Servlets + JSP (embedded Tomcat)	8082

Features

Students

• Build a CV profile (education, experience, skills, interests) and export it as PDF
• Browse and apply to internship postings
• Get AI-powered internship recommendations based on their profile (Gemini)
• Log work hours / journal entries during an active internship
• Track application status and evaluation results

Companies

• Post and manage internship listings
• Review student applications and CVs, accept or reject candidates
• Review and comment on student work logs
• Submit end-of-internship evaluations

Faculty

• Manage company and student records (including bulk student import via CSV)
• Oversee internships, applications, and work logs across the system
• Review evaluations submitted by companies

Architecture

• Backend — Spring Boot REST API with a modular, domain-driven structure (student, company, internship, application, evaluation, worklog, cv, recommendation, user). Stateless JWT authentication, PostgreSQL with Flyway migrations, OpenAPI/Swagger documentation, CSV and PDF export, and an AI client abstraction (Gemini, with a stub implementation for local development without an API key).
• student-app — a single-page application (React Router, typed API client) consuming the backend REST API directly.
• company-app / faculty-app — server-rendered Java web apps (Servlets + JSP) that act as thin clients, calling the backend through a typed gateway layer rather than hitting the database directly.
• Infrastructure — PostgreSQL 16 and pgAdmin, orchestrated via Docker Compose for local development.

Tech Stack

• Backend: Java, Spring Boot (Web, Security, Data JPA, Validation, WebFlux), PostgreSQL, Flyway, JWT (jjwt), springdoc-openapi, OpenPDF, Apache Commons CSV, Google GenAI (Gemini)
• Student App: React 18, TypeScript, Vite, React Router
• Company / Faculty Apps: Java, Jakarta Servlet API, JSP, embedded Tomcat
• Infrastructure: Docker Compose, PostgreSQL, pgAdmin

Getting Started

Prerequisites

• Java 17+ and Maven
• Node.js and npm
• Docker (for the PostgreSQL database)

Run everything

The repo includes a run.bat script that starts the database, backend, and all three client apps in one go:

run.bat

This will:

1. Start PostgreSQL (and pgAdmin) via Docker Compose
2. Start the backend on http://localhost:8080
3. Start company-app on http://localhost:8081
4. Start faculty-app on http://localhost:8082
5. Start student-app on http://localhost:5173

Manual setup

# 1. Start the database
docker compose -f backend/docker-compose.yml up -d

# 2. Start the backend
cd backend && mvn spring-boot:run

# 3. Start each client app in its own terminal
cd company-app && run-dev.bat   # or run-dev.ps1
cd faculty-app && run-dev.bat
cd student-app && npm install && npm run dev

Before running, set real values for security.jwt.secret and ai.api-key in backend/src/main/resources/application.properties (placeholder values are provided for local development). If no Gemini API key is configured, the recommendation service falls back to a stub implementation.

API Documentation

Once the backend is running, interactive API docs are available at:

http://localhost:8080/swagger-ui

Project Structure

.
├── backend/          # Spring Boot REST API
├── company-app/       # Server-rendered app for companies
├── faculty-app/        # Server-rendered app for faculty staff
├── student-app/       # React + TypeScript SPA for students
├── docker/            # Docker Compose setup for local infra
└── run.bat            # One-command startup script (Windows)

Author

Vedran Topić — github.com/vedrantopicc