Hi, I'm Touati Kamel 👋
AI & Data Science Engineer @ESTIN | NLP, LLMs, CNNs | Full Stack Developer
TK

About

I am an AI & Data Science Engineer and a graduate of the HIGHER SCHOOL OF COMPUTER SCIENCE AND DIGITAL TECHNOLOGIES - BEJAIA, where I earned my degree with honors (mention Tres Bien). I have a solid foundation in full-stack web development and a growing expertise in artificial intelligence. Over the past few years, I've led and contributed to impactful projects from building dynamic web applications using modern frameworks, to integrating machine learning models into real-world applications. Passionate about data-driven solutions, I actively participate in Kaggle competitions. My current focus lies in advanced AI domains including Convolutional Neural Networks (CNNs), Natural Language Processing (NLP), and Large Language Models (LLMs). I thrive in collaborative, problem-solving environments and continuously seek opportunities to bridge the gap between software engineering and artificial intelligence to create intelligent and user centric solutions.

Work Experience

F

Freelance

2023 - Present
Full-Stack Developer
Designed, developed, and deployed full-stack web platforms for various clients, specializing in creating data-driven applications with modern technology stacks.
E

End-of-Studies Internship
unsupervised Learning
Computer Vision

February 2025 - September 2025
End-of-Studies Internship - Engineering Thesis Project (AI-BASED SYSTEM FOR FOREST FIRE PREDICTION AND DETECTION)
AI system to address critical data-dependency bottlenecks in forest fire management. Key responsibilities and achievements include:Developed an unsupervised machine learning ensemble(K-Means, GMM, Isolation Forest) in Python to accurately predict daily fire risk from raw meteorological data, successfully eliminating the dependency on scarce historical fire records. Engineered an innovative auto-labeling pipeline using a pre-trained segmentation model and OpenCV to systematically generate a large-scale, multi-class annotated dataset, solving the manual annotation bottleneck for visual detection.Trained and fine-tuned a state-of-the-art YOLOv8s, YOLOv10n, YOLOv11n, YOLOv12n models, real-time fire and smoke detection while effectively distinguishing them from environmental confounders like fog and industrial smoke to minimize false alarms. Built and deployed 'ForstsGuard', a full-stack, multilingual web application with a Python Flask backend and a responsive frontend, providing a practical proof-of-concept and an accessible interface for both the prediction and detection AI modules.
I

Internship
next.js
express.js
redux toolkit

January 2025 - March 2025
Internship (Full-Stack Developer) - ESTINUP incubator Website
Developed the ESTINUP incubator website using Next.js, React, and Tailwind CSS for frontend, and Node.js with Express.js for backend. Implemented authentication with NextAuth.js, managed state using Redux Toolkit, and integrated MongoDB for data storage. The website facilitates interactions between students, teachers, and entrepreneurs, supporting innovative projects in AI and Cybersecurity.

Skills

React
Next.js
Typescript
Node.js
ُExpress.js
ُDjango
ُFlask
ُFast api
Mysql
Mongodb
Postgres
Tailwind
Bootstrap
Sass
jQuery
Javascript
Postman
Python
C
Java
pytorch
tensorflow
sklearn
pandas
numpy
matplotlib
My Projects

Check out my latest work

I've worked on a variety of projects.

ForstsGuard: AI-Powered Forest Fire Prediction & Detection

ForstsGuard: AI-Powered Forest Fire Prediction & Detection

My end-of-studies engineering thesis project. ForstsGuard is an integrated, dual-component AI system designed to overcome critical data-dependency bottlenecks in forest fire management. The system features an unsupervised ML ensemble for predicting fire risk without historical data and an innovative auto-labeling pipeline to train a state-of-the-art YOLOv12n detection model. Both modules are integrated into a full-stack, multilingual web application, providing a practical tool for proactive risk assessment and incident detection.

Python
PyTorch
Scikit-learn
Flask
OpenCV
Computer Vision
Deep Learning
Unsupervised Learning
jpal-dz (Pêche project) - freelance project

jpal-dz (Pêche project) - freelance project

Peche is an innovative app designed for both professional and amateur fishermen in Algeria. It allows users to easily track and manage their fishing history, including catches, locations, and dates. Whether you're a beginner or an experienced angler, Peche helps document your adventures and optimize your trips. For administrators, it provides an advanced management interface with detailed reporting and statistical analysis, offering a comprehensive overview of fishing activities nationwide. This facilitates resource management and informed decision-making. Download Peche today for streamlined and efficient fishing management in Algeria!

Next.js
Typescript
TailwindCSS
axios
SIPTHAL - Information System for Fisheries Traceability - freelance project

SIPTHAL - Information System for Fisheries Traceability - freelance project

integrated platform designed to ensure the recording, monitoring, and traceability of fishery products for first-salemarkets in Algeria's fish auction halls. The system provides comprehensive tracking capabilities for the fishing industry,helping maintain product quality and supply chain transparency from catch to market

Next.js
Typescript
TailwindCSS
axios
Laptech-Final-Version

Laptech-Final-Version

LapTech is an e-commerce platform dedicated to the Algerian market, specializing in buying and selling laptops. This project incorporates cutting-edge technologies, including an AI model, to enhance the user experience by providing accurate price estimates based on laptop specifications.

HTML5
TailwindCSS
React
Redux
Node.js
Express
MongoDB
Postman
chatbot

chatbot

A chatbot built for interacting with users and providing automated responses based on queries.

HTML5
CSS3
JavaScript
Bootstrap
Django
Ai-project-1CS-

Ai-project-1CS-

An AI project developed as part of a first-year computer science course, integrating Flask and Bootstrap.

HTML5
CSS3
JavaScript
Bootstrap
Flask
Library-management-system-project

Library-management-system-project

A robust system for managing library books, members, and transactions with user-friendly interfaces.

HTML5
CSS3
JavaScript
Bootstrap
jQuery
Django
TechnoBlog

TechnoBlog

A modern blog platform built with React, Redux, and Node.js, featuring a clean UI and robust backend.

HTML5
TailwindCSS
React
Redux
Node.js
Express
MongoDB
Postman
Ecommerce-website-project

Ecommerce-website-project

A full-featured e-commerce website for managing products, orders, and customer information.

HTML5
CSS3
JavaScript
Bootstrap
jQuery
Django
Certificates

Professional Credentials

Courses and certifications I've completed to expand my knowledge.

DeepLearning.AI logo

Supervised Machine Learning: Regression and Classification

DeepLearning.AI

March 14, 2024

DeepLearning.AI logo

Advanced Learning Algorithms

DeepLearning.AI

April 6, 2024

DeepLearning.AI logo

Unsupervised Learning, Recommenders, Reinforcement Learning

DeepLearning.AI

April 12, 2024

Hackathons

I like building things

During my time in university, I attended 4+ hackathons and datathons. People from around the country would come together and build incredible things in 2-3 days.

  • N

    NASA Space Apps Hackathon

    Estin School

  • I

    InnoByte Hackathon

    Syphax Hotel

  • H

    Hackathon Bejaia

    Raya Hotel

  • T

    TRAINIT Datathon

    Estin School

Contact

Get in Touch

Want to chat? Just shoot me a dm with a direct question on Linkedin and I'll respond whenever I can. I will ignore all soliciting.