My Projects
A showcase of my academic and personal projects in computer science, machine learning, and web development
Soccer - Computer Vision and Player Tracking
This project develops an advanced computer vision system for real-time soccer analysis, combining YOLO-based player detection, multi-object tracking, and field mapping to monitor player movements and assess in-game danger levels. The system identifies high-threat zones, evolving passing opportunities, and changes in team shape during attacks. Beyond sports analytics, it advances AI research in real-time perception, spatial reasoning, and context-aware decision modeling, with techniques applicable to autonomous systems, robotics, and video understanding.
Revenue Prediction - Orange Money
I developed a complete predictive analytics system to forecast the monthly Revenue of multiple telecom services at Orange Côte d’Ivoire. The project involved building a scalable forecasting pipeline using Facebook Prophet, enriched with domain-specific regressors, seasonal patterns, national holidays, and historical growth signals. I engineered a custom module, PredictionCAUtils, responsible for preparing features, generating future datasets, and producing accurate prediction ranges for services such as CANAL and E-Commerce. The system was designed to integrate seamlessly with operational workflows and provide actionable decision-support for commercial strategy.
Inspection Drone – Thermal Overheating Detection
This project presents an intelligent drone-based inspection system that uses thermal imagery and computer vision to detect overheating in telecom infrastructure. By training a YOLOv8 model on thermal data, the system identifies critical components on pylons and analyzes their temperature levels. A custom thermal color–to–temperature mapping estimates global scene temperature, while a deviation-based diagnostic rule compares individual components to their surroundings to flag potential failures. The solution enables automated, accurate, and efficient monitoring of infrastructure health.
Genora – RAG-Powered AI Assistant for Human Resources
Genora is an AI-powered HR assistant built on a Retrieval-Augmented Generation (RAG) architecture to support employees and streamline internal HR communication. It combines natural language understanding with a document retrieval system to deliver accurate, policy-based answers drawn from official HR documents. Using vector databases, document chunking, and an LLM reasoning engine, Genora provides fast, consistent, and context-aware responses to common HR questions, reducing HR team workload while improving accessibility and the employee experience.
Face Mask Detection through Image classification
This project uses deep learning to improve preparedness for future public health emergencies. It focuses on automatically detecting whether individuals are wearing a face mask correctly, incorrectly, or not at all. Built with convolutional neural networks (CNNs) using TensorFlow and Keras, and supported by OpenCV for real-time image processing, the system can be deployed in public monitoring environments such as cameras or kiosks. Through effective data preprocessing and augmentation in Python, the model achieves robust performance across different conditions. Overall, the project highlights how AI can support health authorities by reducing transmission risks and strengthening early-response strategies during future outbreaks.