Adam Saidane

Adam
Saidane

Building systems that scale,
models that predict, products that matter.

Second-year Software Engineering student at INSAT. I build full-stack web applications, data science pipelines, and intelligent systems — from REST APIs to advanced AI models.

Engineer by training,
builder by nature.

I'm Adam, a Software Engineering student at INSAT in Tunis. My work sits at the intersection of software architecture, data systems, and AI, and I bring that range into every project I build.

During my internship at IID, I designed and delivered production-ready banking APIs using Spring Boot, pushed test coverage beyond 96% with JUnit and Mockito, and led a full CRM version migration in a live codebase.

Beyond that, I build projects driven by curiosity: a reinforcement learning traffic system that reduced waiting times by 60%, a football Expected Goals model using XGBoost, and a full-stack HR management platform.

I'm trilingual — Arabic, English, French — and I enjoy working in collaborative, international environments on challenging, high-impact problems.

Since Sept. 2023

INSAT — National Institute of Applied Sciences and Technology

Software Engineering

Active

Sept. 2019 – June 2023

LPM8 — Menzah 8 Pioneer High School

Baccalaureate · Mathematics

Highest Honors

Where I've shipped things.

International Information Development (IID)

Analyst Developer Intern

July – August 2025 · Tunis

Designed and implemented Java Spring Boot REST services for the ECHO Credit Card project (card lifecycle management) and ECHOS Ecommerce (financial calculation services).

Developed robust unit tests using JUnit and Mockito, achieving 90%+ code coverage across banking APIs — mocking external dependencies, validating edge cases, and ensuring reliable error handling.

Managed a full version migration of the Edeal CRM platform, resolving post-migration bugs and implementing custom business logic for sales automation.

Spring Boot JUnit Mockito Swagger REST APIs

Things I built.

PROJECT_01

AI Traffic Lights Manager

Research-grade adaptive traffic signal control using Deep Reinforcement Learning on a 4-way intersection via SUMO/TraCI. Implements three algorithms — tabular Q-Learning, Double Dueling DQN with Prioritised Replay, and PPO with entropy decay — operating on a 73-dimensional state vector. Best result: DQN achieves ~60% reduction in average waiting time vs fixed-time signals.

PythonPyTorchQ-LearningDQNPPOSUMO/TraCITensorBoard

GitHub

PROJECT_02

xG Model — Football Analytics

End-to-end Expected Goals pipeline built on StatsBomb open data. Separate XGBoost models for each shot category (foot/head/penalty/1v1), enriched with spatial and contextual features including goalkeeper positioning and defensive pressure. Includes an interactive Streamlit app where you click the pitch to get a live xG prediction.

PythonXGBoostStreamlitmplsoccerScikit-learnStatsBomb

GitHub

PROJECT_03

HR Management System

Enterprise-grade HRMS built with ASP.NET Core 8. Covers the full employee lifecycle: onboarding, payroll, recruitment pipeline, promotion tracking, and equipment management. Features a real-time dashboard with 40+ analytics views, three-tier RBAC (AdminRH / Manager / Employee), and cookie-based auth with Claims authorization.

ASP.NET Core 8C#EF CoreSQL ServerRazor Views

GitHub

PROJECT_04

LivresRares — Rare Books E-Commerce

Full-stack Symfony e-commerce platform for rare books. Complete catalogue with search/filter/pagination, AJAX cart with real-time updates, checkout flow with order confirmation, customer review system, and a full admin interface with CRUD for products, users, orders, and categories.

Symfony 6PHPTwigMySQLDoctrineDocker

GitHub

PROJECT_05

Takwira API — Football Backend

RESTful Spring Boot API for football team management. Full CRUD for players, matches, and goals, JWT authentication, role-based access control, Swagger/OpenAPI docs, and Docker + Docker Compose deployment.

Java 17Spring BootJWTMySQLDockerSwagger

GitHub

Code I gave back.

Keras — Deep Learning Library

Contributor

Implemented input validation for PReLU and LeakyReLU activation layers.

Diagnosed a backend-specific failure in OpenVINO during testing, isolating it as a pre-existing issue.

keras-team/keras ↗

What I've earned.

IBM 2026

IBM AI Engineering Professional Certificate

Comprehensive professional certification covering machine learning, deep learning, neural networks, computer vision, and NLP. Includes hands-on projects using TensorFlow, Keras, PyTorch, and Scikit-Learn.

Machine Learning Deep Learning TensorFlow PyTorch Keras

Harvard / CS50 2024

CS50x — Introduction to Computer Science

Harvard's renowned introduction to computer science and programming. Covers algorithms, data structures, memory management, web development, and problem-solving in C, Python, SQL, and JavaScript.

Algorithms C Python SQL Web

Adam
Saidane

Engineer by training,
builder by nature.

Where I've shipped things.